Drone360 Menu
X

SEARCH SITE

Enter keywords or a search phrase below:

News & Notes

Drones Help Improve Wildlife Research

Drones have major potential in the world of wildlife research, but questions and challenges still remain

September 12, 2016

Math can be deadly for wildlife biologists. Researchers and managers quantify animal populations, model migration patterns, and calculate birth and death rates. This means spending much of their time simply counting the animals they study.

Wild animals are elusive, often living and breeding in habitats that are near impossible to reach on foot — human feet, at least. To sidestep the terrestrial trials of the wilderness, not to mention the bugs, claws, bared teeth, and venom, biologists traditionally strap into small airplanes and helicopters to perform their counts. However, ask any scientist who has conducted aerial surveys and they’ll tell you that using small manned aircraft for research is where the trouble truly begins.

Light aircraft crashes were the leading cause of death for wildlife workers between 1937 and 2000, according to D. Blake Sasse’s 2003 analysis published in the Wildlife Society Bulletin. Of the 91 field deaths he verified, 60 were caused by airplane or helicopter accidents. A single fatal snakebite was the only animal-caused death documented by Sasse.

Surveying wildlife from the air is challenging: Pilots fly at low speeds and altitudes, increasing the risk for collisions with power lines and tall trees. During survey flights, pilots also make frequent sharp turns, increasing the chances of a stall as they sweep back and forth across the sky to thoroughly investigate an area.

“Every biologist knows somebody who’s died in a plane or helicopter crash,” says David Bird, professor emeritus of wildlife biology at McGill University, who, coincidentally, studies birds, specializing in predatory raptors.

In 2007, Bird started incorporating drones into his wildlife research. Today, he’s among the leading wildlife researchers in North America expanding the applications for drones in the wild. Bird has led studies using drones to observe osprey, great horned owls, caribou, geese, and more. Now retired, he and his research partner Paul Pace, a former military man with 30 years’ experience flying UAVs, team up with scientists around the world, deploying drones to assist with their work.

Bird, who says he always loves talking about drones, is also the founding editor of the two-year-old Journal of Unmanned Vehicle Systems. In April, Bird compiled a special digital issue of the journal, highlighting ways UAVs are aiding wildlife researchers and managers. The wide array of research from Bird’s journal and other studies from the field offers a glimpse at how drones are faring in the wild.

“Not a day goes by where I don’t hear about someone using drones in a new way to study wildlife,” Bird says.
White pelican counts are cheaper, safer, and more efficient using drones — and, as a bonus, produce some stunning aerial photographs of the ocean birds.
USFWS
Avian overwatch

Drones equipped with high-resolution cameras excel at two primary field operations: overflights and close-up flights. Fixed-wing drones that can stay aloft for hours and snap tens of thousands of images along preprogrammed flight-paths are ideal for overflights.

Once the photos are captured, the images are stitched together and counted manually. Marine mammals and water birds have been the most frequent survey subjects in UAV-aided studies.

In a study published this past March in Nature Scientific Reports, Jarrod Hodgson, an ecologist at the University of Adelaide in Australia, compared the accuracy of UAV-derived counts to those taken by on-the-ground observers of lesser frigatebirds, royal penguins, and crested terns in Australia. He found that UAV counts were slightly larger — due to duplicate counts and misidentified birds — than those recorded on land, but there was also less variance in inaccuracies. In other words, the drones were wrong, but they were predictably wrong. This uniform margin of error is more easily accounted for in statistical models and trends.

Orien Richmond, a biologist with the U.S. Fish and Wildlife Service, has organized UAV-aided water bird surveys around the country and has a pretty good idea of what capabilities drones can bring to the table.

“The range of small UAV systems is pretty limited, and you have to keep them within visual line-of-sight. It’s not practical for carrying out large-scale surveys — at least not right now,” says Richmond.

For example, drones aren’t so useful for a mid-winter waterfowl survey of migrating birds, which typically needs to cover an area the size of the San Francisco Bay in a couple of days. Instead, drones excel in smaller areas, like the pelican nesting site Richmond studied at the Anaho Island National Wildlife Refuge in Nevada. “They do shine for colonial nesting bird studies,” he explains.

To date, there isn’t reliable software that can automatically detect and count wildlife species in images, but Bird believes that such software would be a beneficial addition to wildlife research.
There she blows! Researchers used a hexacopter to capture up-close, high-definition photographs of a Northern Resident Community killer whale pod in 2014. NOAA Fisheries, Vancouver Aquarium
Drive-by drone flying

Close-up flights are tailor-made for multirotor UAS, coming in handy when an animal’s location is known beforehand and researchers can fly relatively close. Bird has come to rely on multirotors to study predatory bird nests constructed high in tree branches or along precarious cliff sides. In the past, Bird would either have needed to hire someone to climb a 50-foot-tall tree or shell out serious cash to rent a helicopter.

While studying osprey nests in Montana, Bird and a fellow researcher used a drone to outwit the hyper-vigilant ospreys. Super aggressive, ospreys will start attacking at just the sound of a drone spooling up its motors on the ground.

“Goshawks, great horned owls, and ospreys are extremely aggressive and will attack anything that comes near their nests,” says Bird.

So Bird and his team hovered the drone beneath the nest, lured the osprey out, and swiftly maneuvered the drone over the treetops to snap a quick photo of the nest’s contents. They did this for 80 different nests, logging successes — for the most part. “In one nest check, an osprey wheeled around and smacked our $2,000 drone to the ground,” he says.

Researchers used a similar drive-by approach to study killer whales of the Northern Resident Community around Vancouver Island off the coastline of
British Columbia. J.W. Durban, a biologist with the National Oceanic and Atmospheric Administration, and his team steered a boat near the whales’ location. When they closed in, they launched a drone off the deck to snap images of the aquatic mammals. They conducted 60 such flights, each lasting about 13 minutes — all while steering the boat to keep the drone within visual line-of-sight.

In total, they captured over 18,000 images with their hexacopter. They used the images to identify individual whales from unique markings and determine if they were in good health based on length estimates. Previously, Durban’s team studied Southern Resident Community killer whales from a helicopter, and their first application using drones at sea proved to be a successful, safe alternative.
Drones to the rescue

While researching animals with drones is all well and good, wildlife biologists would have nothing to research without the animals themselves. So conservation efforts are in the biologists’ (and the animals’) best interests.

Drones have proved integral to anti-poaching efforts in Africa. They help monitor animals, track poacher movements, and even intimidate criminals who might want to press their luck by searching for illegal trophies in a national park. Mixing drones with heavily armed poachers presents its own dangers, but even more basic conservation measures can still be harrowing.

It’s estimated that collisions with nearly-invisible power lines kill hundreds of thousands of birds annually. Clipping on small, reflective line markers that increase the visibility of the wires is a solution that reduces collisions by up to 50% in some areas. But installing the little reflectors is ridiculously complicated. Typically, a pilot flies a helicopter within 3 to 5 feet of a wire while a technician sits — or more precisely, dangles — outside the helicopter to secure the marker.

“You’re brave if it works out, foolish otherwise,” explains James Dwyer, an environmental scientist at the Fort Collins, CO-based utilities consulting company EDM International. “It’s pretty involved, because there are only a handful of companies that do this, and those companies only have a handful people that can fly these operations.”

These marker-placing crews are in high demand, and their dockets fill up pretty quickly, so power companies that want to make their lines safer need to wait ... in line.
In a proof-of-concept study, Dwyer assisted a team of scientists using drones to attach markers on a mock power line. They attached an extended arm to a hexacopter and fitted it with a marker that snapped into place once it contacted with the wire. They flew the drone up to the line, snapped on a marker, landed the drone, and repeated the process.

“Once we have things in place, we are installing at a similar pace to what you would on a helicopter. We would need to take frequent breaks to switch batteries, however,” says Dwyer.
No need to risk broken limbs to peek into a bald eagle nest — drones are able to maneuver more safely for both the birds and the researchers. James Junda
The demonstration is one of many ways drones are being applied in more creative ways for conservation purposes. At the U.K.’s University of Exeter, for instance, Andrew Cunliffe and his team used drones in 2015 to monitor how recently reintroduced beavers affected the landscape of the River Tamar in south west England. They photographed the sites — without mucking through wetlands — and used data processing to generate orthophotos (aerial photos corrected for scale) and digital surface models. With this, they quantified how beavers improved waterflow and reduced flooding.

“Sharing that story with members of the wider public, farming communities, and policymakers makes the visual products from drones very valuable, in addition to the hardcore quantification they help with,” says Cunliffe.

Diving beneath the surface, researchers at Queensland University of Technology in Australia are hoping their autonomous, starfish-killing underwater drone called COTSbot will join the fight to save the Great Barrier Reef. The underwater ROV autonomously seeks and destroys crown-of-thorns starfish, a predatory scourge that feasts on coral. The COTsbot injects the ravenous starfish with a fatal cocktail of vinegar using a metal proboscis.

Bird and Pace are also experimenting with drones equipped with thermal imaging cameras to help a farmer spot bobolink blackbird nests in his field. Bobolinks breed primarily in open hay fields, so the farmer wanted to be sure he could avoid their nests while baling.

And on the more adventurous side of things, Swiss biologist Roman Eyholzer plans to see if drones fitted with tranquilizer guns could be used to subdue wolves — he only recently received the grant money to begin the experiment, but research should be underway soon.
Using a basic digital camera mounted to a drone, researchers can model how beavers alter the landscape. Andrew Cunliffe/University of Exeter
Don’t be bullish

There are still questions that researchers need to answer before UAVs become ubiquitous in the field. For starters, scientists are just beginning to understand how drones affect the various animals they research. Many studies anecdotally mention whether the drones appeared to disrupt wildlife — it seems drones don’t bug most animals. But only a handful of studies have attempted to test the question in a controlled way.

Mark Ditmer, a biologist at the University of Minnesota Twin Cities, found that flying drones near wild black bears fitted with heart rate monitors caused as much as a 400% spike in heart rate — a strong indicator of stress or excitement. Ditmer’s study is one of the most well-known and well-cited attempts to quantify drones’ effects on wildlife.

He’s conducting a follow-up study this summer to see if captive black bears also show signs of stress in the presence of drones, and if they habituate to the technology. Ditmer will test this by repeatedly flying a UAV over bears at the Wildlife Science Center in Columbus, MN.

“It’s a way to pump the brakes a little bit,” says Ditmer. “We need to develop some best scientific practices before we can dive into all the applications that drones provide. There are so many papers coming out that say drones will revolutionize conservation and wildlife research, and we agree with that, but there are questions to answer.”

In a similar study, researchers in France used a quadcopter to fly over populations of mallards, flamingos, and common greenshanks. They found that 80% of flights didn’t seem to have a measurable impact on the birds.

However, the birds reacted more when drones approached vertically, because that flight pattern resembles the behavior of a predator. Researchers recommended launching drones more than 300 feet from the birds, adjusting approach distance and angle according to species.

Meeting birds and other animals on their terms is essential to the safety and data integrity of a research operation. Sending a colony of nesting waterbirds into flight could cause them to kick their eggs and young over a cliff — a high price to pay if you’re working with endangered species.
On the wing

Eventually, scientists will home in on best practices and figure out how to best use their drones in the field. But that’s not the biggest barrier for the advancement of UAV-assisted wildlife studies.

Most developed nations prohibit UAV operations beyond visual line-of-sight (BVLOS), and Bird and others hope those regulations will be eased for researchers, not hobbyists. Part 107 enables researchers to file for BVLOS waivers.

“That’s the big one. It’s what we’re all waiting for,” says Bird. “Until that happens, we really can’t do the types of things we’d like to with drones.”
We’re quickly moving into an era where drones will be able to detect obstacles and fly independently, which makes the idea of autonomous wildlife telemetry quite alluring.

Fixed-wing drones could autonomously fly over a large swath of land, probing for radio signals blinking from a wide-ranging creature like a tagged polar bear. “Such a solution could be revolutionary,” Bird wrote in a review article he published in the Journal of Unmanned Vehicle Systems. The idea is certainly compelling, but it hasn’t moved past the theoretical simulation stage.

“I can’t think of many situations going forward where beyond line-of-sight isn’t a big deal,” says Ditmer. “We got a lot less flights in [during our black bear study], simply because of where the animal was located. We knew [the drone] would immediately leave our line-of-sight, and we collected a lot less data.”
Bird says an underlying reason for launching the Journal of Unmanned Vehicle Systems was to showcase the great things drones are doing for society. He emphasizes that this is a journal for everyone, not just researchers.

The information the journal contains could ultimately allay the general public’s concerns about drones and their impact on privacy and safety, encouraging more acceptance.

Though he knows government will move slowly, he remains confident that policy and technological advancements will continue to progress, ensuring that
wildlife researchers can use these tools to improve the lives of not only humans, but also our furry, feathered, or scaled animal neighbors.

Note: A version of this story appeared in the September/October 2016 issue of Drone360 magazine.
Featured image: USFWS