A comparison of three methods for planning a census of Tawny Owl (Strix aluco) populations living at high territorial density

Abstract

Censusing nocturnal species such as the Tawny Owl (Strix aluco) living in a dense population can lead to an overestimate if individual signatures are not available. A technique that separates the individualistic call of the Tawny Owl could be a useful tool for an exhaustive census. Six Tawny Owl males were followed for nine months; 654 vocalizations were analysed. Three methods – Spectrographic Cross Correlation (SPCC), Acoustic Space (AS) and Visual Spectrogram Comparison (VSC) – were tested to assess their ability to classify the typical male call. For SPCC, 10 randomly selected hoots for each male were compared: the distributions of correlation coefficients differed in only 26.7% of the cases when intra and inter individual variability are compared. For AS, all the spectrograms were measured through 13 parameters and intra and inter individual distances were compared: the interval containing 95.4% of intra-individual measures also contained 95.9% of inter-individual comparisons. Both SPCC and AS are considered not to be able to adequately separate subjects. For VSC, 31 randomly selected hoots were visually compared by 5 helpers; their classifications were compared pairwise and with the real situation; operators correctly identified a male in 70% of cases (mean = 70.4 SD = 5.4). If we integrate VSC with information coming from field notes, we have a more powerful tool than the mapping method. It is plausible that this technique can be useful for an exhaustive census of Tawny Owl populations living at high densities.