Ambient Noise in an Urbanized Tidal Channel

In coastal environments, when topographic and bathymetric constrictions are combined with large tidal amplitudes, strong currents (> 2 m/s) can occur. Because such environments are relatively rare and difficult to study, until recently, they have received little attention from the scientific community. However, in recent years, interest in developing tidal hydrokinetic power projects in these environments has motivated studies to improve this understanding. In order to support an analysis of the acoustic effects of tidal power generation, a multi-year study was conducted at a proposed project site in Puget Sound (WA) are analyzed at a site where peak currents exceeded 3.5 m/s. From these analyses, three noise sources are shown to dominate the observed variability in ambient noise between 0.02-30 kHz: anthropogenic noise from vessel traffic, sediment-generated noise during periods of strong currents, and flow-noise resulting from turbulence advected over the hydrophones. To assess the contribution of vessel traffic noise, one calendar year of Automatic Identification System (AIS) ship-traffic data was paired with hydrophone recordings. The study region included inland waters of the Salish Sea within a 20 km radius of the hydrophone deployment site in northern Admiralty Inlet. The variability in spectra and hourly, daily, and monthly ambient noise statistics for unweighted broadband and M-weighted sound pressure levels is driven largely by vessel traffic. Within the one-year study period, at least one AIS transmitting vessel is present in the study area 90% of the time and over 1,363 unique vessels are recorded. A noise budget for vessels equipped with AIS transponders identifies cargo ships, tugs, and passenger vessels as the largest contributors to noise levels. A simple model to predict received levels at the site based on an incoherent summation of noise from different vessel types yields a cumulative probability density function of broadband sound pressure levels that shows good agreement with 85% of the temporal data. Bed stresses associated with currents can produce propagating ambient noise by mobilizing sediments. The strength of the tidal currents in northern Admiralty Inlet produces bed stresses in excess of 20 Pa. Significant increases in noise levels at frequencies from 4-30 kHz, with more modest increases noted from 1-4 kHz, are attributed to mobilized sediments. Sediment-generated noise during strong currents masks background noise from other sources, including vessel traffic. Inversions of the acoustic spectra for equivalent grain sizes are consistent with qualitative observations of the seabed composition. Bed stress calculations using log layer, Reynolds stress, and inertial dissipation techniques generally agree well and are used to estimate the shear stresses at which noise levels increase for different grain sizes. Ambient noise levels in one-third octave bands with center frequencies from 1 kHz to 25 kHz are dominated by sediment-generated noise and can be accurately predicted using the near-bed current velocity above a critical threshold. When turbulence is advected over a pressure sensitive transducer, the turbulent pressure fluctuations can be measured as noise, though these pressure fluctuations are not propagating sound and should not be interpreted as ambient noise. Based on measurements in both Admiralty Inlet, Puget Sound and the Chacao Channel, Chile, two models are developed for flow-noise. The first model combined measurements of mean current velocities and turbulence and agrees well with data from both sites. The second model uses scaling arguments to model the flow-noise based solely on the mean current velocity. This model agrees well with the data from the Chacao Channel but performs poorly in Admiralty Inlet, a difference attributed to differences turbulence production mechanisms. At both sites, the spectral slope of flow noise follows a f-3.2 dependence, suggesting partial cancellation of the pressure fluctuations when the turbulent scales are on order of, or smaller than, the characteristic size of the hydrophone. At both sites, flow-noise levels can exceed ambient noise levels during slack currents by more than 50 dB at 20 Hz and flow-noise is measured at frequencies greater than 500 Hz. In Admiralty Inlet, the use of a compact flow shield is shown to reduce flow-noise levels by up to 30 dB. Below 1 kHz, the dominant source of ambient noise is vessel traffic, though during periods of strong currents, the propagating noise from vessels can be difficult to identify because of flow-noise. At frequencies above 1 kHz, during periods of strong currents, the dominant source of ambient noise is bedload transport. Observation of this higher frequency sound is not affected by flow-noise, which is limited to lower frequencies in northern Admiralty Inlet. These results are combined with marine species hearing thresholds, a turbine source spectrum, and a simple propagation model to roughly quantify the probability of marine animals detecting the sound of operating turbines against ambient noise. The results suggest that the likely detection range of operating turbines is limited to less than 1 km under most conditions. The sound produced by operating tidal turbines at the proposed demonstration-scale tidal power project is not likely to have any significant behavioral effect at greater range. Finally, the ambient statistics at the site are also combined with a sound propagation model and vocalization characteristics of Southern Resident killer whales to determine the effective range for passive acoustic monitoring techniques at the proposed project location. Due to the frequency overlap between sediment-generated noise and killer whale vocalizations, during peak currents the detection range for vocalizations is reduced by up to 90% when compared to slack current noise levels. Although the reduction in detection range is significant, this analysis suggests that passive acoustic monitoring will still be effective at ranges greater than the typical range at which killer whales can detect the turbines. These results of these two detection studies will inform the design of post-installation monitoring plans to quantify noise production by operating turbines and the associated environmental changes. This dissertation provides a comprehensive analysis of ambient noise measurements in an energetic coastal environment and advances the understanding of noise sources unique to these environments, such as sediment-generated noise and flow-noise. The improved understanding of these noise sources will aid in the interpretation of acoustic measurements in other energetic environments. Furthermore, as uncertainties in sound produced by tidal turbines and marine animal behavioral responses to this sound are reduced, the foundation laid by this research will allow the acoustic impacts of tidal hydrokinetic power projects to be quantified.