Research Project Finds Meaning in Birdsong
Cadet Sarah Hunziker ’13 spent last summer recording and cataloging the songs of tree swallows in an effort to understand the meanings behind their vocalizations.
The research was conducted under the mentorship of biology professor Col. Dick Rowe and sponsored by the Swope Summer Scholars Program, which consists of eight weeks of intensive research. Hunziker, who is considering graduate studies in marine biology, is currently completing the analysis of the birdsongs for her capstone project.
“I know that if I go to graduate school, field research is something I’m very interested in doing,” said Hunziker. “It was a really good experience and was really eye opening to what my life could be like for the next few years.”
The research was conducted at the biology department’s new field site at the Sky Farm area in McKethan Park, where birdhouses were set in the spring to attract the migrating tree swallows. The tree swallows were present by the summer, and a high-sensitivity microphone was used to record the swallows’ songs.
“We were trying to get as close as we could to the birds in order to isolate just one bird and record its vocalizations,” said Rowe. “By paying attention to what the birds were doing when they were vocalizing, we can start to understand what those songs mean.”
All of those calls were carefully cataloged along with the various contexts in which they were made.
“I took extensive field notes, and the recorders allow you to mark different parts of the tape,” said Hunziker. “I could mark the tape and then say in my notes ‘I was x amount of feet from the nest box and they were dive bombing me and the recorder,’ so I would pair up the recordings with the context.”
By documenting the context along with the recordings, researchers can begin to identify meaning behind the songs. Using sound analysis software helped to visualize and quantify minute differences in the calls.
“There’s a huge amount of information in these calls. It’s very subtle, and that’s where the software comes in,” said Rowe. By comparing tiny differences in the duration, amplitude, and pauses, researchers can find and quantify variations in the songs that may not be readily noticeable by the human ear.
“We can look at it like a sentence: notes are like letters, syllables are like words, and a song is like a sentence,” said Rowe. “We suspect that there are some really complicated things going on.”
To get the best data, research was conducted at sunrise, when the birds are most active. Since the site was so close to post, four or five trips a week were possible for the entirety of the two-month research period. All that time in the field created a mountain of data to be analyzed.
“For every minute you spend in the field there are ten or so on the computer: playing it back, sorting through it, and cutting out background noise and other birds that I don’t need,” said Hunziker.
From those hours of recording, Hunziker isolated what she suspected were alarm calls and strung them together in one recording. Playing that recording back to the swallows triggered aggressive flight patterns.
“Within 30 seconds we had tree swallows circling and diving at the tape recorder trying to figure out where the threat was,” said Rowe. “It was very clear that this was an alarm call and that there was some kind of recruitment piece in there.”
Now that the alarm calls have been identified and documented, the next round of research will seek to determine what kind of information is contained within those alarm calls.
Recent research has established that the alarm calls of both mammals and birds are referential; that is, they are able to convey information about the nature of the threat. Referential alarm calls refer to types of threats and will differ depending on the predator that triggered the calls. These referential calls are opposed to risk-based calls, which simply relate to the proximity or intensity of the threat.
“It’s possible that tree swallows just have risk-based alarm calls,” said Rowe. “On the other hand, they may be saying, we have a snake, we have an owl, or we have a bluebird problem.”
Future research projects on this subject would seek to determine whether tree swallow calls are referential by setting up scenarios and monitoring the birds’ responses.
“If they give a certain call in response to a hawk, and we play that song back to them in the absence of a hawk, do they act as if a hawk is around?” said Rowe. “So you have to set up behavioral scenarios to correlate behaviors with a song.”
The migratory swallows have now vacated McKethan Park for the winter but are expected back in time for next summer’s research.
–John Robertson IV