| Summary: | Thesis (Ph.D.)--University of Washington, 2016-12 This thesis focuses on fin whale vocalizations recorded on ocean bottom seismometers (OBSs) in the Northeast Pacific Ocean, using data collected between 2003 and 2013. OBSs are a valuable, and largely untapped resource for the passive acoustic monitoring of large baleen whales. This dissertation is divided into three parts, each of which uses the recordings of fin whale vocalizations to better understand their calling behaviors and distributions. The first study describes the development of a technique to extract source levels of fin whale vocalizations from OBS recordings. Source levels were estimated using data collected on a network of eight OBSs in the Northeast Pacific Ocean. The acoustic pressure levels measured at the instruments were adjusted for the propagation path between the calling whales and the instruments using the call location and estimating losses along the acoustic travel path. A total of 1241 calls were used to estimate an average source level of 189 +/-5.8 dB re 1uPa @ 1m. This variability is largely attributed to uncertainties in the horizontal and vertical position of the fin whale at the time of each call, and the effect of these uncertainties on subsequent calculations. The second study describes a semi-automated method for obtaining horizontal ranges to vocalizing fin whales using the timing and relative amplitude of multipath arrivals. A matched filter is used to detect fin whale calls and pick the relative times and amplitudes of multipath arrivals. Ray-based propagation models are used to predict multipath times and amplitudes as function of range. Because the direct and first multiple arrivals are not always observed, three hypotheses for the paths of the observed arrivals are considered; the solution is the hypothesis and range that optimizes the fit to the data. Ray-theoretical amplitudes are not accurate and solutions are improved by determining amplitudes from the observations using a bootstrap method. Data from ocean bottom ...
|
|---|