Improving individual identification of wolves (Canis lupus) using the fundamental frequency and amplitude of their howls : a new survey method

摘要: Many bioacoustic studies have been able to identify individual mammals from variations in the fundamental frequency (F0) of their vocalizations. Other characteristics vocalization which encode individuality, such as amplitude, are less frequently used because problems with background noise and recording fidelity over distance. In this thesis, I investigate whether inclusion amplitude variables improves accuracy howl identification captive Eastern grey wolves (Canis lupus lycaon). also explore use a bespoke code extract features, combined histogram-derived principal component analysis (PCA) values, can improve current wolf accuracies. From total 89 solo howls six individuals, where distances between observer were short, achieved 95.5% (+9.0% improvement) using discriminant function (DFA) classify simple scalar F0 normalized amplitudes. Moreover, was increased 100% when PCA values amplitudes first harmonic.