Study subject: An aerial shot of dugongs captured by the UQ team in Moreton Bay.

Credit: Amanda Hodgson

Related articles

• Antarctic voyage dredges up new species
• Automated killer robots: threat to humanity
• Remote-control phones listen in to rare birds
• Australia plans massive conservation corridor
• Satellites build a picture of the past

Packed into a tiny, low-flying plane, a crew of five marine scientists and their pilot slowly circle a patch of ocean off Queensland, over and over, around and around. Below them, munching away on seagrass, dugongs congregate in their thousands.

The researchers need to work together in the confines of their aircraft to count and clock the position of the animals; but members of the swarming herd all look alike.

Adding to their difficulties, the team is also rushing against the return of a 10-knot wind threatening to impair visibility through the water. It's an expensive hindrance that's already kept them grounded at their remote outpost for the past week.

"It's all a bit challenging," says wildlife biologist Amanda Hodgson, from the University of Queensland in Brisbane, Australia. "When you think about everything that has to go on, conducting [a manned] aerial survey is quite tricky, logistically."

Not to mention dangerous, too. "We've all had scary experiences while out surveying in one of those little planes," Hodgson says. "And yes, there have been accidents, even fatalities." Though none have been in Australia, Hodgson cites accidents, which have resulted in eight deaths, during aerial marine mammal surveys in other parts of the world.

Novel approach

Yet aerial identification and counting of marine mammal populations – particularly vulnerable marine species such as dugongs (Dugong dugon) and whales – provides invaluable data. It allows fluctuating animal numbers to be monitored, and most crucially, Hodgson says, it allows key habitat areas to be identified and targeted for conservation.

With these issues in mind, the University of Queensland (UQ) team are currently testing a different approach to aerial surveying: automated aircraft, known as unmanned aerial vehicles (UAVs).

Starting this month, Hodgson and her teammate Michael Noad, will be among the first in the world to trial the use of UAVs in wildlife surveillance. Their aim is to evaluate the surveying capabilities of a camera-mounted, unmanned flight system.

Their work will focus on counting populations of dugongs and humpback whales (Megaptera novaeangliae) in particular, and involve collaboration with survey and tracking experts from James Cook University in Townsville, Queensland, and the University of Newcastle in New South Wales. It will also involve input from the Queensland-based UAV business, Aerocam Australia.

If all goes well, the researchers believe an unmanned approach to aerial surveying would make studies easier, eliminate the human risk element, and reduce costs. It may also offer improved accuracy, and allow remote and hostile environments to be studied with ease.

A flying shadow

According to Greg Smith, of Aerocam Australia, UAVs have a rich history, stretching back as far as World War I. "They were originally designed for the military to collect intelligence information in hostile environments, without endangering pilots and aircrew," he says.

Now they are increasingly used in the civilian realm, for everything from police and emergency service observations, to the investigation of cyclones in meteorology.

Yet despite their history, the use of UAVs in wildlife surveillance is relatively new. "There's definitely a lot of interest in the field, but there isn't any current, published research out there," Hodgson notes. "Our project is actually one of the very first to directly test UAVs for this kind of application."

The UAV they will be getting familiar with is the three-metre-long Shadow. It boasts a maximum flight range of 1,500 km, equal to eight hours flight time, and will initially be mounted with both cameras and video equipment.

These flying robots can work in a number of different ways says K.C. Wong, a UAV flight systems expert and mechanical engineer at the University of Sydney in New South Wales.

"[They] can operate with varying levels of autonomy," he says, "ranging from having a remote ground-based pilot controlling all aspects of the mission, to a fully autonomous system where the ground-based operator merely monitors… and makes critical decisions."

The Shadow will be flown using a combination of these automated and manual methods. While the initial flight plan will be programmed into the autopilot prior to take off, a live video link will allow the researchers to take over the controls when needed – to fly the plane over a straggling whale, for example, or to cover a large area where dugongs are spread out.

Eye in the sky: The Shadow is a three-metre-long unmanned aerial vehicle (UAV). It can fly with varying degrees of autonomy and has found a new role photographing and counting marine mammals off the coast of Queensland [Credit: University of Queensland / Aerocam Australia]

Variety of techniques

According to UQ's Michael Noad, humpback whales and dugongs are "the best test case animals we have" for evaluating the usefulness of UAV technology. They are both key animals to survey in terms of the importance of their habitat and population conservation. But they also require the use of two very different surveying techniques, which means they'll provide a broad assessment of the capabilities of the unmanned planes.

"Dugongs need to be observed straight down through the water, where they clump together in big herds," says Noad. "But humpback whales are completely different… most of the time they're too deep to see. You can really only look for their blows and breaches [at the surface]."

At first the research team will aim to test-fly their UAV over the marine animals to determine the best flight parameters, such as altitude and speed. It isn't until later in 2008 that the actual surveying will begin in the waters near Brisbane. The plan is for the humpbacks to be counted off North Stradbroke Island in July, and the dugongs to be monitored later in October, in both Moreton Bay and Hervey Bay.

"We really don't know what we'll find," says Noad. "Probably the biggest potential problem with UAVs is how well we're going to be able to see the whales and dugongs," he says. "The human eye is an extraordinary sensor… and no matter how good the light-sensing systems of video and camera are, they won't quite be a replacement for actually being there."

"However what we hope is that [because] UAVs potentially cost so little to put up, we can more than overcompensate for this, just by having them in the air for much longer periods of time," he says. With a manned survey, researchers may cumulatively spend a few tens of hours in the air, but with the UAV, aerial data will be collected over "literally hundreds and hundreds of hours", Noad says.

The use of videos and still images offers the additional advantage of creating a permanent survey record, something that is missing from work with the naked eye, says Hodgson: "We'll have all the sightings documented, so we can reassess them as many times as we need to."

Unlocking UAV potential

What's more, still and video cameras may just be the starting point in unlocking the survey potential of the UAV. "We're trying to be systematic about it, and take one thing at a time," says Noad, "[but] we are still thinking ahead about the different types of sensors we can eventually try and use."

"I think if we can get infrared [cameras] working, an unmanned survey could turn out to be far superior to any manned flight," he says. Infrared imaging would allow the animals to be detected with enhanced accuracy and clarity, and could also allow night time surveying.

"At the end of the day, we're probably a few years away from using UAVs properly in [wildlife surveillance]… even if all goes perfectly well with our study," says Noad. "But their future does hold enormous potential."

Wong agrees. "One can see them being used as an extension of our ability to observe ecology and animals remotely… covering a significantly larger area with minimal disturbance to the natural environment."

But there are also numerous other potential applications for UAVs, he says, such as search and rescue, coastwatch, landmine detection and autonomous cargo delivery, and no doubt a fistful of exciting applications we've yet to even dream of.

Readers' comments