Abstract
Marine hydrokinetic turbines are a promising technology for generating renewable ‘green’ energy at the scale of local communities, as well as for feeding regional power grids. Many locations along Nova Scotia’s Bay of Fundy with significant potential for marine turbine power have prototype turbines either installed or in planning. Developing of these sites requires an understanding of the resource potential and the effects of the turbines on the environment. Typically, desktop models are employed to gain an initial understanding of these issues, which are then supplemented with in situ measurements. The measurements serve two purposes: to validate and to improve the models or to address knowledge gaps identified during the desktop studies. With respect to the effects of turbines on marine life, a key knowledge gap is understanding the behaviour and exposures of fish, crustaceans, and marine mammals near the devices.
To help address these knowledge gaps, Environmental Effects Monitoring Plans (EEMPs) for installing and operating prototype marine tidal turbines often include requirements for passive acoustic monitoring (PAM). PAM may be employed to detect the presence of vocalizing marine life or to quantify the sound levels in a project area with and without the tidal turbines. The sound level measurements may then be used to determine how turbines change the sound exposures of marine life, including the possibility of sounds from the turbine causing injury to hearing, behavioural disturbance, or masking of other biologically important sounds. The importance of these possible effects must be interpreted in the context of their cumulative effects with pre-existing human activities.
Measurements made by EEMP programs record the conditions at the sensor locations over the recording period. To convert these into an understanding of the effects over wide areas and long durations, we need to validate that the available modelling tools accurately reproduce the real-world. In the case of PAM, we have access to numerical propagation models whose accuracy depends on our knowledge of the environmental conditions and the sound levels generated by sources in the area. Environmental conditions that matter are the speed of sound in the water column, the seabed topography and composition, and the sea surface conditions. Sites that are appropriate for tidal turbines are known to have high levels of turbulence. As a result, the water column sound speed fluctuates with values on the order of 0.1–1% of the mean value, which has an unknown effect on our ability to model sound propagation. Sound sources whose levels need to be quantified include those of the turbines as well as other activities like fishing vessels and ferries.
The scope of EEMPs developed for specific projects cannot address some of the fundamental research questions, such as acoustic propagation effects. Programs such as those managed by the Offshore Energy Research Association (OERA), are appropriate for addressing these types of questions. Our program “How does sound travel in high energy environments? Effectiveness of acoustic monitoring systems and turbine audibility assessment” was designed to advance our ability to perform and interpret PAM programs. The objectives of the program were to:
- Measure propagation loss for sounds in the 8–16 kHz band;
- Investigate directional sensors as acoustic receivers to localize marine mammals near tidal turbines;
- Measure the baseline ambient soundscape in Grand Passage, including the source level of the Brier Island Ferry;
- Measure propagation loss in lower frequency bands (100–5000 Hz) using the ferry as a sound source;
- Develop a stochastic variability model for acoustic propagation in turbulent conditions; and
- Estimate the audibility zone of turbines for different marine life groups.
An acoustic measurement program lasting a full 28-day tidal cycle was conducted to fulfil these objectives. An acoustic projector provided a controlled signal in the 8000–16000 Hz frequency range for the high-frequency propagation loss study. The projector’s signal was recorded by three acoustic data loggers at nominal ranges of 100, 500, and 1100 m from the source. The 100 and 500 m recorders straddled to the Brier Island Ferry route (Figure 1) so that estimates of propagation loss and ferry source levels could be made.
The acoustic equipment was deployed and retrieved during the previous reporting period. The equipment was deployed on 20 and 21 Sep 2018 and retrieved on 27–28 Oct 2018. Appendix A contains a description of the recording equipment, acoustic source, deployment and retrieval operations, as well as the non-acoustic data that was collected to support data interpretations.
This report provides an overview of the project team, a summary of the results, and recommendations for further work. Details of the work performed are contained in the appendices.