評論
打印
This is Scientific American — 60-Second Science. I'm Karen Hopkin.
Ecologists crouching quietly amidst vegetation, using binoculars to tally birds in a roost, may soon be a charming relic of the past. Because a new study shows that, when it comes to getting an accurate avian head count, aerial drones can do better.
In recent years, scientists who study wild populations are increasingly turning to remotely piloted aircraft...otherwise known as drones...to monitor their animal of interest. For example, drones are being used to track pods of whales...or to keep an eye on African elephant herds and watch for signs of poaching.
Such remote surveys are generally considered highly cost-effective. But it wasn't clear whether they are as accurate as old-fashioned, feet-on-the-ground, expert evaluations.
To find out, researchers in Australia set up a test.
"And so what we did was make some replica seabird colonies where we knew the true number of individuals in each colony."
Jarrod Hodgson of the University of Adelaide led the study.
Using decoy-sized rubber ducks, the researchers laid out 10 colonies...ranging in size from about 500 to more than 1,000 individuals.
"We then had experienced ground counters make independent counts of those birds from nearby, from the optimum vantage point. At the same time, we flew a drone overhead capturing photographs at different heights above the colony."
The drone data were then analyzed two ways. First, a gaggle of citizen–scientists was tasked with adding up the number of birds they could see in each scene.
The results of that approach:
"We found that on average the drone-derived data or the drone-derived counts made by humans counting the images were between 43 percent and 96 percent more accurate than the traditional ground-based counts."
The better the image, the more accurate the count.
Then, Hodgson and his team developed a semi-automated computer program to do the counting for them. And they found:
"...there was no statistical difference between those counts and the counts completed by our volunteers using exactly the same imagery."
The results appear in the journal Methods in Ecology and Evolution.
So counting by drone not only saves time and effort, but yields better data...a definite plus in terms of conservation.
"When we monitor wildlife, increasing the accuracy and the precision of animal surveys gives us more confidence in our population estimates. And this means that we have a stronger evidence base on which to make management decisions or policy changes."
In other words, a drone in the sky is better than two PhDs in the bush. Who can rely on better data to be feathers in their caps.
Thanks for listening for Scientific American — 60-Second Science. I'm Karen Hopkin.
