Abstract

Effective monitoring of rare and declining species is critical to enable their conservation, but can often be difficult due to detectability or survey constraints. However, developments in acoustic recorders are enabling an important new approach for improved monitoring that is especially applicable for long-term studies, and for use in difficult environments or with cryptic species.

Bioacoustic data may be effectively analysed within an occupancy modelling framework, as presence/absence can be determined, and repeated survey events can be accommodated. Hence, both occupancy and detectability estimates can be produced from large, coherent datasets. However, the most effective methods for the practical detection and identification of call data are still far from established. We assessed a novel combination of automated clustering and manual verification to detect and identify heathland bird vocalizations, covering a period of six days at 44 sampling locations.

Occupancy (Ψ) and detectability (p) were modelled for each species, and the best fit models provided values of: nightjar Ψ = 0.684, p = 0.740, Dartford warbler Ψ = 0.449, p = 0.196 and woodlark Ψ = 0.13, p = 0.996. Including environmental covariates within the occupancy models indicated that tree, wetland and heather cover were important variables, particularly influencing detectability.

The protocol used here allowed robust and consistent survey data to be gathered, with limited fieldwork resourcing, allowing population estimates to be generated for the target bird species. The combination of bioacoustics and occupancy modelling can provide a valuable new monitoring approach, allowing population trends to be identified, and the effects of environmental change and site management to be assesse

Link to Journal Website

citation: ‘Abrahams, C., Geary, M., 2020 Combining bioacoustics and occupancy modelling for improved monitoring of rare breeding bird populations. Ecological Indicators (112). https://doi.org/10.1016/j.ecolind.2020.106131’