Our birds are deserting us. In fact, an estimated 50,000,000 birds departed from the northeastern United States on Monday night, when winds were particularly favorable. By next spring, we are going to know exactly where a few of them went.
We’re getting better at studying migration. That’s good, considering that the overall North American population of birds has declined by 3 billion since 1970. We really need to know how many of our birds are disappearing because of conditions here, and how many are dying because of conditions where they go in winter. So the question that must be answered before all others is this: where exactly are they going?
Most tracking is still accomplished by the old tried and true method of banding birds. This is particularly useful for birds that nest in open colonies, where the babies are banded before they can fly. Visit any puffin island and notice how many puffins and terns have bands. Endangered species, such as the piping plovers on Maine’s southern beaches, often get the banding treatment. Peregrine falcon chicks get banded, too.
Songbirds are a little trickier, but banding them also yields a treasure trove of information. Many species travel thousands of miles to the tropics every autumn, then find their way back to the same nesting areas in the spring, where they can be recaptured. It might seem more difficult to recover banded birds when they die en route. However, a disappointingly large number are killed by colliding with skyscrapers and other obstacles, where they can be collected on the pavement at dawn.
Biologists have been radio-tagging birds since the 1960s. It’s been a godsend for tracking large birds like whooping cranes, but the hefty transmitter and battery pack formerly limited its usefulness on smaller birds. The rule of thumb is that the device can’t exceed 5 percent of the bird’s weight. Fortunately, technology is now so miniaturized that even smaller birds can be outfitted with a radio or GPS unit.
Still, these tiny units can be hugely expensive, and battery life is always a challenge. So ingenious scientists have come up with new devices to solve the problem. Among these, nanotags have produced thrilling results. Some are so small, they can be glued onto the backs of butterflies. They’re perfect for tracking seasonal migration, because they last only about a month, and the glue is designed to release the tag soon after the device dies.
Nanotags operate a bit like your E-ZPass on a toll highway. They are transponders that remain inert until triggered by a signal. When triggered, they send back the bits of information that identify that individual bird. Successful tracking requires a system of transmitters along primary migration routes, because the effective range is only about 10 miles. But since many birds follow coastlines, rivers and mountain ridges, that’s often adequate.
Geolocators are wicked clever devices. They neither transmit nor receive. They merely store data. Specifically, they record daily sunrise and sunset. Since the precise occurrence of each varies by latitude and longitude, the data can reveal the location of the bird each day with surprising accuracy.
Geolocators are about the size and weight of a paper clip. Because they require little power and minimal memory, geolocators can track and store a bird’s location over long periods, sometimes years. It’s best used for birds that can be recaptured in the spring, since the device must be recovered and the data downloaded by humans, but they are quite inexpensive compared to telemetry units.
There remains one last problem. How do you identify individual birds you can’t catch and equip? Until now, you couldn’t. When a bunch of finches come to a feeder, the human eye just isn’t able to discern identifiable differences between each one. But it turns out, a computer can. Using the same type of artificial intelligence that enables computers to learn and facially recognize individual people in a crowded airport, scientists can train a computer to recognize individual birds.
Netting and equipping individual birds is time-consuming and expensive, and it’s a bit stressful for the bird. As technological improvements accelerate, it is becoming possible to track and study individual birds without ever touching them. Currently, computers must be fed many photos of known individuals in order to learn how to recognize them. Eventually, they likely will be able to teach themselves how to recognize new birds.
And so we come to mid-September. Birds are on the move, and biologists are on the move with them.
Bob Duchesne serves as vice president of Maine Audubon’s Penobscot Valley Chapter. He developed the Maine Birding Trail, with information at mainebirdingtrail.com. He can be reached at firstname.lastname@example.org.