AUTOMATIC FEATURE EXTRACTION AND CLASSIFICATION OF CROSSBILL (LOXIA SPP.) FLIGHT CALLS

ABSTRACT

In this paper a new method for the automatic classification of bird sounds is presented. Our method is based on acoustic parameters (features) taken from the first harmonic component computed from the sound spectrogram. The features are based on a line segment approximation of the first harmonic component. The final feature vectors, consisting of 16 real numbers, are then classified using a self-organizing map (SOM) neural network. Flight calls of four crossbill species (Loxia spp.) are used as a test example. In the first phase, an unsupervised network was trained and tested using common crossbill L. curvirostra flight calls recorded mainly in the Netherlands. The network was tested using two-barred L. leucoptera, Scottish L. scotica and parrot L. pytyopsittacus crossbill flight calls in the second phase. Finally, the results were validated applying the same network to flight calls of common crossbills and parrot crossbills recorded in Finland. The method automatically separated common crossbill flight calls from those of parrot crossbills. The classification accuracy of the Dutch recordings was 58% in the first phase and 54% in the second phase. The Finnish recordings were classified with 54% accuracy.     

Under a neighbour's influence: public information affects stress hormones and behaviour of a songbird

Socially acquired information improves the accuracy and efficiency of environmental assessments and can increase fitness. Public information may be especially useful during unpredictable food conditions, or for species that depend on resources made less predictable by human disturbance. However, the physiological mechanisms by which direct foraging assessments and public information are integrated to affect behaviour remain largely unknown. We tested for potential effects of public information on the behavioural and hormonal response to food reduction by manipulating the social environment of captive red crossbills (Loxia curvirostra). Red crossbills are irruptive migrants that are considered sensitive to changes in food availability and use public information in decision making. Here, we show that public information can attenuate or intensify the release of glucocorticoids (i.e. stress hormones) during food shortage in red crossbills. The observed modulation of corticosterone may therefore be a physiological mechanism linking public information, direct environmental assessments and behavioural change. This mechanism would not only allow for public information to affect individual behaviour, but might also facilitate group decision making by bringing group members into more similar physiological states. The results further suggest that stressors affecting entire populations may be magnified in individual physiology through social interactions.     

Coping with the extremes: stress physiology varies between winter and summer in breeding opportunists

Seasonal changes in stress steroid hormone secretions are thought to reflect investment in self-maintenance versus reproduction. The capricious conditions hypothesis (CCH) posits that reduced corticosterone (CORT) secretion during stress coincident with parental phases of breeding is necessary in harsh environments because a full response would otherwise trigger repeated nest abandonments. To test this hypothesis, we measured seasonal changes in stress physiology in free-living red crossbills (Loxia curvirostra), an opportunistically breeding songbird that regularly breeds in summer and winter. This species allows unique comparisons of breeding physiology under very different seasonal environmental conditions within locations. We found strong support for the CCH: red crossbills showed reduced CORT secretion only when in high reproductive condition in the winter, when compared with summer breeders and winter non-breeders. These data demonstrate that behavioural status and local environmental conditions interact to affect mechanisms underlying investment trade-offs, presumably in a way that maximizes lifetime reproductive success.     

On the relative importance of ecology and geographic isolation as drivers for differentiation of call types of red crossbill Loxia curvirostra in the Palearctic

Efforts to understand the process of speciation have been central to the research of biologists since the origin of evolutionary biology as a discipline. While it is well established that geographic isolation has played a key role in many speciation events, particularly in birds, there is ongoing debate about how frequent speciation is in the partial or complete absence of geographical isolation. In the red crossbill Loxia curvirostra, good arguments do exist for sympatric speciation processes. In this species, several classes of calls are clustered in distinct groups, so-called ‘call types', which mate assortatively. Often, several call types can be found at a single site, breeding and feeding next to each other. It has been hypothesized that red crossbill call types evolved by specialising in extracting seeds from cones of different conifer species. Alternatively, call types might have evolved in temporal geographic isolation. Within Europe, little is known about the distribution of the various call types and preferences for distinct food resources. In this study, we analysed the temporal and spatial occurrence of red crossbill call types in the Palearctic, investigated potential reasons for call-type composition at a site, and compared the occurrence of call types with the fructification of conifers. Call-type composition changed with site and season but hardly with conifer species. With our data, we could localise range areas of twelve different call types, which cannot be explained by conifer species occurrence. Therefore, we suggest that call types evolved in parapatry in most of the northern Palearctic region, and, although contradictory results exist from Iberia, we argue that differentiation might be driven by the same drivers there as well. Additionally, we discuss the potential influence of anthropogenic changes of forest composition and distribution on call types, which offers a unique possibility for future studies.