Dr. David C. Lahti
Queens College, City University of New York
Our project on weaverbird song, which began in Sept 2016 and is set to end in June 2018 (pre-publication) as per our grant application, is proceeding well. However, we have had to move portions of our work around due to two sorts of constraints-- one in the field and one in the lab. In the field, breeding was delayed at first due to odd weather, but we have been recording wonderfully since January. We know have hundreds of good songs, which are waiting for analysis.
The songs were to be at least pre-analyzed by this month, however, the software we use in the lab is specialized for tonal songs, and is performing poorly on weaver song that has not only diphony (two notes being sung at the same time), but also a lot of harmonics and broadband elements (clicks, rattles). We knew weaver song would be complex and difficult to quantify, but it was anyone's guess how we would solve this problem once we got songs in the lab. I have decided to switch software to Sound Analysis Pro (Ofer Tchernichovski), as that was designed for zebra finch songs and so can deal well with these elements. However, to get a student in Ofer's lab learning how to use that software will have to wait until this September. In the meantime recording and data manipulation will continue, although there's very little we can say about the songs themselves until we can analyze them.
The globular nests and complex rambling songs of Ploceus weaverbirds are striking and familiar features of the sub-Saharan African landscape. The village weaver (Ploceus cucullatus), the most abundant member of the genus, breeds in colonies where the cacophony of their simultaneous singing can hardly be overlooked along gallery forests and waterways, around agricultural fields, and within villages and towns. Precisely because of their coloniality and the fact that much of their courtship and breeding interactions occur within the colony, individual village weaver songs are rarely recorded. We consequently have little understanding of how these songs vary between individuals, over geographic distances, or between species. Moreover, weaverbirds often change their behavior in response to human intrusion, decreasing the opportunity for researchers to record their song, and potentially rendering them different than if they had been recorded during ordinary activity. Because of the Wildlife Acoustics grant of a SongMeter SM4, we have been able to record weaverbirds individually, and without having to be physically present so the focal birds are not alarmed but sing and behave normally. We have also been able to record begging calls from young birds, which we would otherwise never have been able to do in a colonial species in a natural context.
Since we received the Wildlife Acoustics SongMeter SM4 in 2016 for this project, our recordist Clive Barlow has made a variety of excellent recordings of the whole range of Gambian village weaver vocalizations, from the calls of nestlings, fledglings, parents, and mates, to the competitive and mate- attractive songs of males of all ages in both breeding and eclipse (nonbreeding) plumage. These recordings are revealing the diversity of vocalization in this species, and will also constitute the first element in a broader comparative study of song across the genus Ploceus. Specifically in the village weaver, the SongMeter SM4 has already provided us unprecedented access to two important vocal features: individual song structure, and the relationship between the begging calls of weavers and the cuckoo chicks that parasitize their nests.
The complex songs of the village weaver are consistently delivered as a series of phrases of multiple types, which in succession give a distinct impression of a rise in intensity. This impression is corroborated by analysis of amplitude (volume), frequency (pitch), bandwith (pitch range), and note rate, all of which tend to increase across comparable elements in the course of a complete song. The phrases are each either extended or repeated indeterminately, and might at any juncture either lose steam and peter out or else build and transition to a predictable next phrase type in the series. Often the full songs can be divided into a chirpy introduction, two or three variations on a warble-trill-buzz theme, and a trilled coda. The following recording illustrates two renditions back-to-back of the song of a typical Gambian adult male, recorded by the SM4 in June 2016. A series of call-like notes build to a complex warble that ends in a long flat buzz; then comes a brief warble, trill, and a second buzz, this one rising; this is followed by a brief warble and trill and a descending whistly buzz; the song ends with an extended rapid trill.
The diederik cuckoo (Chrysococcyx caprius) is a brood parasite of the village weaver, meaning that its adult females lay eggs in weaver nests rather than laying in nests of their own. A successfully parasitized weaver raises none of its own young, but only a cuckoo chick, during that reproductive attemptâthe cuckoo mother and (eventually) the chick remove any weaver eggs and nestlings from the nest. One might expect cuckoo chicks either to mimic weaver chicks in their food begging calls, or simply to deliver more engaging calls in some way to compensate for the fact that they are not otherwise similar in appearance to young weavers, especially as they grow older and beg for food outside of the nest. Recordings by the SM4 of both weaver and cuckoo juveniles begging from weaver parents (or foster parents) so far show a similar rate of delivery, but no evidence of structural mimicry. Cuckoo begging calls appear to be more tonal (melodic) than weaver calls at a given age.
Figure 3. Village weaver male song, recorded on the SM4, The Gambia (C. Barlow, recordist).
The following audio tracks correspond to the recording in Figure 3. Track A is at normal speed. Track B is slowed to one-quarter speed to show the acoustic detail, which in some respects also more faithfully represents the listening experience of a songbird, given how quickly they process audio input and distinguish rapidly delivered notes.