Critical ratios of beluga whales (Delphinapterus leucas) and masked signal duration

Abstract

This article examines the masking of a complex beluga vocalization by natural and anthropogenic noise. The call consisted of six 150ms pulses exhibiting spectral peaks between 800Hz and 8kHz. Comparing the spectra and spectrograms of the call and noises at detection threshold showed that the animal did not hear the entire call at threshold. It only heard parts of the call in frequency and time. From the masked hearing thresholds in broadband continuous noises, critical ratios were computed. Fletcher critical bands were narrower than either 15 or 111 of an octave at the low frequencies of the call (<2kHz), depending on which frequency the animal cued on. From the masked hearing thresholds in intermittent noises, the audible signal duration at detection threshold was computed. The intermittent noises differed in gap length, gap number, and masking, but the total audible signal duration at threshold was the same: 660ms. This observation supports a multiple-looks model. The two amplitude modulated noises exhibited weaker masking than the unmodulated noises hinting at a comodulation masking release.