May 05, 2017
One state wildlife agency is using drones, some with thermal imaging, in an elk survey project.
Text and images from Oregon Dept. of Fish and Wildlife
By Oregon Dept. of Fish and Wildlife
ASTORIA, Oregon â€“ From their vantage point high atop the Oregon Coast Range, Oregon Department of Fish and Wildlife biologist Herman Biederbeck and two researchers from Oregon State University can see almost forever as the first rays of sunlight peek over the top of Saddle Mountain in the distance to the east.
Below is the Young's River basin and a patchwork of thousands of acres forest land interspersed with clear-cuts â€“ ideal elk habitat.
The researchers, Jonathan Burnett and Cory Garms, both Ph.D. students in the Department of Forest Engineering, Resources and Management at OSU, want to find out whether unmanned aerial systems (UAS) or "drones" can be used effectively to count elk in this kind of terrain.
Preliminary results of field trials conducted on the North Coast near Astoria suggests that they can.
"UAS technology has promise to be relatively inexpensive and safe â€“ much safer â€“ than the way we survey elk now, which is generally from a helicopter," said Biederbeck, a wildlife biologist with ODFW for 38 years.
This year's field trial in Clatsop County is the first time that UAS technology has been used to count elk in Oregon, although ODFW has used drones to survey salmon spawning in rivers and as well as cormorant abundance along the Oregon coast.
ODFW conducts yearly elk population/composition surveys to make sure that age and sex ratios stay healthy.
"It's part of our mission to monitor these populations to ensure they are being well managed for the public," said Biederbeck.
This year drones were used in two field trials, one in January and another in March. The first tested the drone camera's ability to capture imagery that allows biologists to classify elk by age and sex. A later field trial tested the aircraft's ability to measure elk densities in forest stand types, another useful metric for managing elk.
ODFW currently contracts helicopters at a cost of $1,000 to $1,100 an hour to do this job. The agency staffs them with ODFW employees who look for and document elk in flights conducted year after year over the same survey units for statistical accuracy.
MANNED VS. UNMANNED AIRCRAFT
Each aerial system has its advantages and disadvantages, according to Biederbeck, who notes that with a helicopter, observers can view great expanses of landscape in real time by scanning their eyes in front and to the sides of the airship. Crew members can also ask the pilot to reposition the machine for a better look at animals, which can be especially helpful when it comes to distinguishing elk calves from adults. In addition, helicopters are much heavier and more powerful than drones and can fly in a wider range of weather conditions. The down side is unless they have a hand-held camera on board, observers only get one chance to classify elk â€“ right then and there.
In addition to their relatively low cost, drones have the advantage of recording images that can be reviewed on a computer back at the office. Human safety is one major benefit of the UAS. People can get hurt or even killed in a helicopter. For example, two ODFW biologists, Holly Huchko and Eric Himmelreich, suffered broken bones but fortunately survived a helicopter crash a few years ago while conducting fish surveys on the Umpqua River in southern Oregon.
The drones used in this year's experiment on the North Coast cost about $1,700 apiece, according to Burnett, although the thermal sensor adds another $3,500 to the cost of the system.
As darkness gives way to dawn, the first of two drones is prepared for flight. It is jet black in color, with flashing red night lights on the sides, and thermal imaging equipment on board. Its job is to detect elk hidden in the trees by keying in on their heat signatures with a heat-sensitive infrared camera.
A second drone â€“ white, and equipped with a high definition video camera â€“ will fly as soon as the black one gets back from its mission. The video camera is mounted on a gimbal that lets the drone operator tilt, turn, and pan the camera with a joystick that can also steer the aircraft.
After a turn at the end of one run along the serpentine-shaped run, the camera swivels from pointed directly at the ground to straight ahead toward the next GPS waypoint. The recording is set to overlap video from each pass so the video from each stretch can be "stitched together" with imaging software to so that every inch of the survey area is pictured.
The drones can fly essentially the same survey areas as helicopter in a single flight, according to Biederbeck, but likely take more passes because cameras do not have the same field of view as humans, who are able to scan the whole horizon and turn quickly from side to side with a simple twist or turn of the head.
With takeoff just minutes away, Burnett double-checks the flight path glowing from a laptop in the back of his SUV. A yellow line on the computer screen shows the exact course the aircraft will follow, a series of switchbacks. The route is made by programming GPS coordinates into the drone's navigation system ahead of time.
REGULATORY BARRIERS REMAIN
Each flight lasts about 30 minutes, and the drone follows GPS coordinates automatically, although the pilot can override the navigation software to assume control the vehicle manually. FAA rules require a designated spotter be present and maintain visual contact with the aircraft throughout the flight. The aircraft are battery-powered and are programmed to return to base automatically whenever they detect their batteries are getting low.
This technology is a potentially powerful tool for conducting scientific inquiry, according to Burnett, although many regulatory barriers to effective implementation remain, notably Federal Aviation Administration (FAA) rules.
"Throughout this study there have been two major regulatory limitations to assessing the true cost-benefit of using UAS for elk survey," said Burnett. One limitation is the current 400-foot altitude ceiling. The other is the requirement to maintain line of sight on the aircraft during its flight.
Higher altitudes and greater coverage area on each flight would translate to fewer flights and lower odds of counting the same animals more than once, according to Burnett.
"This technology demonstration is one small step in bridging the gap between what we currently can do and what we ultimately want to do," he said.
Biederbeck and Burnett expect to extend this research by seeking FAA waivers and perhaps acquiring a fixed-wing UAS with up to three-hour flight endurance that may be equipped with both thermal and color cameras.
"There is more operational technology out there. We'll have to see how costs and FAA regulations affect our ability to use them," said Biederbeck.