Localizing Bowhead whales in the Chukchi Sea using asynchronous hydrophones

Abstract

This paper localizes bowhead whales using Bayesian inversion of the modal dispersion of whale calls recorded on asynchronous hydrophones in the northeastern Chukchi Sea, Alaska. A cluster of seven asynchronous ocean-bottom hydrophones (OBH), separated by up to 7.5 km, recorded low-frequency bowhead whale calls as the whales migrated through the Chukchi Sea. The calls dispersed into multiple modes after propagating through the shallow water environment. Relative mode arrival times are extracted from the recordings using a warping time-frequency analysis for nine frequency-modulated whale calls, each of which were recorded on multiple OBHs. A trans-dimensional Bayesian inversion approach is used to invert mode arrival times for the whale location in the horizontal plane, source instantaneous frequency (IF), water sound-speed profile, seabed geoacoustic parameters, relative recorder clock drifts, and residual error standard deviation, all with estimated uncertainties. A simulation study found that accurate localization results could be obtained even without accurate prior environmental knowledge. Inverting multiple calls jointly is shown to significantly reduce uncertainties for whale location, source IF, and relative clock drift. Whale location uncertainties are less than 160 m and clock drift uncertainty is less than 26 ms. This clock synchronization accuracy is sufficient for localizing other types of marine mammal calls using simpler methods (e.g., time-difference-of-arrival).