Song rates of male songbirds can serve as key indicators of territory quality for females. So what happens when intensive selective logging alters these rates? Rajeev Pillay and colleagues from the University of Florida and Universiti Malaysia Sabah summarise their recent research.

*Update November 2019: following the publication of this blog post, the grey-headed canary-flycatcher photo below was selected as our November 2019 cover image.

Selective logging is a major cause of tropical forest degradation and is expanding globally. Research on the effects of selective logging on tropical biodiversity has predominantly focused on changes in population- and community-level parameters that provide useful information on population status, for example abundance and species richness. In comparison, little is known about the effects of selective logging on animal behaviours. Yet many behaviours can be critical for population and community dynamics, while also serving as indicators of habitat quality. Logging-induced adjustments in breeding behaviours such as the song rates of individual songbirds may serve as a potential behavioural indicator of changes in territory quality.

In our study, we used bioacoustic technology to quantify individuals, songs and duets for a community of 32 species of breeding songbirds in the tropical forests of Sabah, Malaysian Borneo. We then applied N-mixture models to these count data to assess the effects of logging on songbird populations (occupancy, abundance) and singing behaviours (per-capita song rates, per-pair duet rates).

Species that showed lower occupancy and abundance in logged forests, relative to unlogged forests, also produced fewer songs per capita and duets per pair. Conversely, species that showed higher occupancy and abundance in logged forests also tended to produce more songs per capita and duets per pair. Furthermore, we found a key ecological trait that determined whether a species showed larger or smaller population size – as well as produced more or fewer songs per capita – with logging, was the extent to which it preferred undisturbed habitats. Thus, species with a high degree of preference for undisturbed habitats (i.e. the species most vulnerable to human disturbances) responded the most negatively to logging in both populations and singing behaviour. In contrast, species known to exploit degraded habitats not only showed larger population size in logged forests but also produced more songs per capita and duets per pair. Given that song rate of individual male songbirds often serves as a cue of territory quality to females, these results indicate that logging may alter territory quality for breeding songbirds, either enhancing it or degrading it, depending on the habitat preferences of species.

This research shows the potential utility of per-capita song rate as a behavioural indicator of changes in territory quality with selective logging. The rich content in bioacoustic recordings can not only facilitate the estimation of important population parameters corrected for imperfect detection, but also enables investigation of otherwise cryptic behaviours relevant to individual fitness and population persistence.

Read the full article, Bioacoustic monitoring reveals shifts in breeding songbird populations and singing behaviour with selective logging in tropical forests in Journal of Applied Ecology.