Applications of Unmanned Aerial Systems (UAS) to Wildlife Research


Small unmanned vehicle systems (UVS), formerly exclusive to militaries, are rapidly advancing in sophistication and availability to civilians. Ranging from hand-launched autonomous airplanes to terrestrial robots to underwater machines, they are increasingly being employed in such areas as agriculture, emergency services, meteorology, oceanography, geophysics and film-making. Another area that shows great potential for the applications of UVS and associated markets is wildlife research and management.

Applications being carried out in various regions of the world today include monitoring breeding, wintering and migrating populations of colonially nesting birds, spawning salmon and orangutans, mapping breeding habitat of endangered species, tracking threatened caribou and polar bears in the far north, examining nest contents of raptorial birds breeding in inaccessible locations, and deterring poachers in Africa. As technology and industry continue to develop and the regulatory procedures begin to loosen, we anticipate an ever-widening range of applications to surface in this field. Some examples include underwater tracking of foraging aquatic birds, detecting signals from songbirds bearing radio-transmitters, surveying nest contents of underground burrows of birds and mammals, and dispersing nuisance birds.

Our current research attempts to explore and promote possible applications of UVS of all types and sizes to help meet the needs of today’s wildlife biologists and managers and to delve into the limitations faced by those wishing to utilize UVS in terms of costs, size, practicality in the field, regulations, etc. We have been investigating the applications of Unmanned Aerial Systems (UAS) to wildlife research for the last six years at McGill University in Montreal. Currently we have three main research thrusts, which cover both birds and mammals in the field. Our first study entails flying a AI Multi fixed-wing UAS over a common tern (Sternus hirundo) colony in New Brunswick to examine spatial distribution and associated habitat use by the nesting birds. We have also developed a less intrusive approach to censusing their nests. In a similar vein, we also mapped habitat use by threatened least bitterns (Ixobrychus exilis) in a military compound in southern Quebec. A second major study in its second season entails using a small rotary-winged UAV to survey the contents of raptor nests, specifically Swainson’s hawks (Buteo swainsonii) and ferruginous rough-legged hawks (B. regalis) in Saskatchewan and osprey (Pandion halieatus) in Montana, as well as to gauge the behavioural responses of the nesting adults to the UAS. To assess the disturbance, we are measuring two variations of parental nest-defense response by documenting the behavior with an ethogram: 1) interspecific variation between raptor species and 2) variation of nest defense behavior within a single nesting cycle in osprey, i.e. eggs, nestlings under 3 weeks, and nestlings over 4 weeks. In 2012 we initiated a study to investigate the efficiencies and limitations of a UAS to locate radio-tagged woodland caribou (Rangifer tarandus), and acquire ‘real-time’ imagery of herds in comparison to using a manned helicopter. We are using the Brican TD100E fixed wing UAS, equipped with an onboard Sigma-eight vhf/GPS tracking system and a camera, to perform flights on the military training grounds in Goose Bay, Labrador. Unmanned aircraft systems could potentially be used to develop a standardized method for conducting population surveys and simplify the process for locating radio-tagged animals.

Specific Ongoing Research Projects

  1. The Rise of Unmanned Aircraft in Wildlife Research and Management: Their Application to the Study of Waterbirds and Beyond

    Dominique Chabot and David M. Bird, McGill University, Ste Anne de Bellevue, QC

    Over the last several years, there has been increasing interest in putting the unique capabilities of unmanned aircraft to use in wildlife research and management, though efforts to date have largely been preliminary in nature or lacking clear context. The aim of this Ph.D. project was to help guide and stimulate the adoption of unmanned aircraft in the field of wildlife research and management using a contextualized and systematic approach. To this end, two case studies were undertaken in which the unique combination of high-resolution imaging and low-disturbance operation by a small fixed-wing unmanned aircraft provided key contributions to full-fledged, management-driven studies of waterbirds.

    In the first study, the aircraft was used to gather fine-scale imagery of a large wetland in Baie-du-febvre, QC, which has been designated by the Canadian Wildlife Service (CWS) as critical habitat for the threatened Least bittern (Ixobrychus exilis), a small and secretive heron. These data played a crucial role in developing a model of bittern habitat preferences within the wetland and observing habitat changes over a multi-year period. The study’s conclusions will now enable the CWS to draft an informed long-term recovery strategy for bitterns in the wetland.

    The second study centred on Canada’s largest Common tern (Sterna hirundo) breeding colony—home to 15% of the continental population, or upwards of 7,000 nests—situated on coastal barrier islands in Kouchibouguac National Park, NB. In order for Parks Canada to improve its colony management plan, a better understanding was required of the terns’ habitat relationships on the highly dynamic barrier islands. In addition, there was interest in developing a less intrusive method of performing annual population censuses than having ground surveyors walk through the entire colony, as terns are notorious for relentlessly mobbing intruders and the beach grass habitat is vulnerable to trampling. The unmanned aircraft proved exceptionally useful towards both objectives as it was able to fly over the colony at low altitudes without startling terns off their nests. Incubating terns were clearly visible in the aerial imagery, allowing a population census to be accomplished and a detailed analysis of nest distribution in relation to habitat to be carried out.

    These two case studies demonstrate how unmanned aircraft can be successfully and beneficially used in conjunction with traditional data collection methods and established GIS tools to achieve real-world wildlife research and management objectives. The final component of this project involves conducting a comprehensive, systematic review of the wildlife scientific literature in order to identify further specific opportunities for unmanned aircraft to make contributions throughout the broader discipline.

  2. Application of Unmanned Aerial Systems to Estimating Caribou Populations in Labrador

    Charla Patterson and David M. Bird, McGill University, Montreal, QC, William Koski, LGL Environmental Research Associates, King City, ON, Brian McLuckie, Brican, Mississauga, ON, and Paul Pace, Carleton University, Ottawa ON

    In southern Labrador and north-eastern Quebec, forest-dwelling caribou (Rangifer tarandus caribou) are deemed to be non-migratory and have adapted to a particular environment and landscape. As of 2002, all sedentary caribou in Labrador were designated as Threatened under the Endangered Species Act of Newfoundland and Labrador. Some of the herds described above occasionally venture into the military training area, e.g. bomb test range, associated with Canadian Forces Base Goose Bay.

    The impact of the aircraft noise involved with such training maneuvers on the health, welfare, behavior, and general ecology of these animals is not fully understood. While roughly 10 percent of all of the caribou in each of the four herds are wearing radio collars with satellite capabilities and thus, provide their monthly locations to scientists, it is not known whether the satellite signals represent the presence of the entire herd or merely the animals bearing the collars. Thus, it would be beneficial to obtain “real-time” imagery of these radio-tracked caribou to discern these possibilities. While manned helicopters offer some unique advantages such as the ability to hover over and land in rugged terrain for closer examination, allowance of in-flight changes to survey plans, carrying several pairs of “eyes” for spotting animals, etc., they are also noisy, environmentally unfriendly in their use of fossil fuels, and very expensive to operate each hour. Most important, each time biologists engage in helicopter surveys, they place their lives at risk.

    As an alternative to manned aircraft for studying caribou populations in Labrador, we are undertaking studies of using a small unmanned aerial system (UAS) to produce accurate aerial surveys and provide “real-time” imagery of herds of caribou in Arctic and sub-Arctic environments. Our goals are to acquire caribou signatures in their environment by means of a manned helicopter, use the UAS to fly a systematic grid and analyze the VHF/GPS receiver data to find the caribou and evaluate its efficiency, fly the UAS in a specified flight pattern over the located caribou and capture digital images of the area, evaluate the level of detail the digital imagery provides, and lay the groundwork for future explorative studies. We look upon the Labrador study as a model for research on the use of UASs for surveying caribou herds such that the technology can be used by other biologists censusing large terrestrial mammals around the world.

  3. Use of Rotary Unmanned Aerial Vehicles (UAV) to Determine Nest Contents of Raptorial Birds

    James Junda and David M. Bird, McGill University, QC; Erick Greene, University of Montana, Missoula

    Conducting surveys of the contents of raptor nests, e.g. eggs, young, has always provided a challenge to investigators. Nests are often in located in precarious or inaccessible locations. Ground-based surveys pose a risk to investigators by way of potentially aggressive parental responses as well as to the nest contents from the disturbance caused by the investigators. While manned light aircraft are still often the tool of choice to check the contents of raptor nests, they are expensive, obtrusive, not very ‘green’, and present potential sampling biases. Most important, manned light aircraft comprise the number one source of mortality for biologists. We are using a rotary-winged Unmanned Aerial Vehicle (UAV) to survey nests of various raptor species with the objective of determining how they respond behaviorally to the approaching machine. In preliminary test flights undertaken in 2011-12 in Missoula, Montana, osprey (Pandion haliaetus) nesting in Montana were highly inclined to attack such machines approaching the nests, but this varied among individual pairs. A preliminary test on an abandoned red-shouldered hawk (Buteo lineatus) nest in a forest with heavy canopy revealed that a UAV equipped with GPS was difficult to maneuver through the branches and moreover, lost its satellite signal due to the canopy. Preliminary flights to one nest each of ferruginous rough-legged hawks (B. regalis) and swainson’s hawks (B. swainsoni) in Saskatchewan in 2012 revealed a total lack of response by the parents. In 2013, roughly 80 flights with a Draganflier X-4 to osprey nests in Missoula resulted in only one strike by an adult. The UAV was slightly damaged, but the bird was unharmed. Flights to the nests of several other raptor species are planned for 2014.

  4. Nuisance Bird Dispersal Using Small Unmanned Aerial Vehicles

    David M. Bird, Avian Science and Conservation Centre, McGill University, 21111 Lakeshore Road, Ste Anne de Bellevue, QC and Paul Pace, Carleton University, Ottawa ON

    In recent years, because of changes in land use, climate changes, and cultural practices, a number of avian species throughout the world have increased their populations to the point of them being labelled as “nuisance birds” (see Sodhi and Sharp. 2006 and references therein). In some cases, birds in large numbers annually cause the loss of hundreds of thousands of dollars to a farmer. For instance, a flock of European starlings (Sturnus vulgaris) can completely clean out 2.5 hectares of grapes in just three days (S. Baptiste, pers. comm.). Just a few other important North American examples include blackbirds (Icteridae) and corn and sunflower crops, snow geese (Chen caerulescens) and winter wheat, and double-crested cormorants (Phalacrocorax auritus) and farmed catfish. Also associated with these exploding bird populations are damage to public and private property as well as increased risks to human health and safety, e.g. airports, town dumps, parks, etc. Such problems are not confined to North America; some international examples include red-billed quelea (Quelea quelea), a flocking songbird that devastates rice crops in several African countries leading to starvation among local villages, and several parrot species, e.g. rosellas, galahs, ravaging ground and orchard crops in Australia.

    In short, the need for control of nuisance birds all over the world is a billion-dollar growth industry! Any means to limit the number of birds in agricultural areas and the associated damages would be of great value. Bird dispersal techniques used in the past and somewhat unsuccessfully today include the use of loud audible alarms, trained raptors and/or dogs, and chemicals or explosives to disperse birds (see Hygnstrom et al. 1994). These devices, however, have proven to have limited effectiveness over time and other techniques which cause physical, debilitating harm and even mortality to the birds, e.g. shooting, poisons, are deemed undesirable by the general public. With the above in mind, the overall aim of our proposal is to to develop the use of a small unmanned aerial vehicle (SUAV) to disperse nuisance birds without causing lasting harm to them and which is publicly acceptable. We are aware of past efforts to use radio-controlled model airplanes that resemble raptorial birds at airports, as well as similar UAVs available today on the internet that purport to scare away birds, but to our knowledge, these have been largely unsuccessful because the birds simply habituate to them over time. We feel that for UAVs to be truly effective in dispersing nuisance birds, the latter have to be made to feel uncomfortable, i.e. have an experience of an unpleasant nature, while present in a spatial location undesirable to humans. Our research objective is to work with Transport Canada to develop a UAS that will effectively disperse nuisance birds.

  5. Rotary and Static Unmanned Aerial Systems as Tools for Radio-tracking Song Sparrows

    David M. Bird, Avian Science and Conservation Centre, McGill University, QC, Paul Pace Carleton University, Ottawa, ON, Liana Zanette, University of Western Ontario, London, ON, Michael Clinchy, University of Victoria, BC, and Afzal Suleman, Centre for Aerospace Research, University of Victoria, BC

    Experiments on song sparrows (Melospiza melodia) on the Gulf Islands have been conducted for the past 12 years to test the effects of predators on the physiology, behaviour and demography of prey. Last year Zanette and Clinchy published a paper in Science (2011, 334:1398-1401) demonstrating that perceived predation risk (i.e. fear) itself is powerful enough to cause a 40 % reduction in the number of young fledged per breeding season. They now want to test whether fear itself also affects survival from fledging to first breeding.

    In the summer of 2012, they conducted a pilot study to test a newly developed radio transmitter from Lotek (www.lotek.com) that can be programmed to turn itself on and off, and so transmit for up to 1 year. They attached tags to 20 song sparrow fledglings, 4 on Russell Island and 16 on Portland Island. These tags are programmed to turn on during the first 10 days of December (2012), and then the first 10 days of April (2013). This past July, they found 17 of the 20 tags. All individuals were still present on the islands on which they had been tagged.

    By December of each year, some of the tagged individuals leave the island on which they were tagged and thus, the sparrow team consequently has to expand their search area. They have a 17-foot Zodiac Pro-12 rigid hull inflatable and so can easily travel to the coastline of all the surrounding islands. Their intention is to conduct transects along the shore of all the surrounding islands checking for the signals of any missing tags. Finding a means to elevate their antenna is essential to improve their signal detection area.

    Some form of Unmanned Aerial System (UAS) might offer a solution to elevating the antenna high enough in the sky over the Gulf Islands to pick up the signals of the radio-tracked birds. Professors Zanette and Clinchy are very interested in collaborative testing of a rotary UAS or a static UAS in the form of a dirigible or even a large kite to be employed to first determine the presence or absence of their tagged song sparrow fledglings on the surrounding Gulf Islands and then finally, to actually triangulate the birds’ position on each of the two islands they nest on.

    The Centre for Aerospace Research (CAR) at the University of Victoria has on hand a wide array of UAS, including two different rotary machines, i.e. a quadcopter and a more powerful hexacopter. CAR also has access to the static models of UAS from Dr. Paul Pace, a Senior Scientist with the Defence Research and Development Canada in Ottawa. For the latter UAS, Dr. Pace possesses the skill and knowledge for using this technology and thus, his presence during the research is paramount. Thus, it will be necessary to provide for his travel expenses from Ottawa to the Victoria area. The idea is to test the concept by “planting” radio-tags in known positions on the islands and testing out the range. Then, we would do so again in the following year when the Lotek tags are due to turn on.