Abstract
A key challenge facing the global marine renewable energy sector is the ability to effectively answer the critical question of the safety of in-steam tidal energy turbines for fish, a key component of the marine environment. Traditional fish sampling technologies, such as trawls, have limited application in high-flow environments. Novel approaches are required to provide the environmental data necessary to achieve public, regulatory, and industry confidence.
FORCE and its partners have been using hydroacoustics to collect information on fish use of the Minas Passage. Two data collection methods have been used: downward-looking, mobile surveys, and upward-looking, stationary surveys. The first method provides spatial coverage of the test site but only spans 24 hours at a time. Conversely, the upward-looking, stationary approach lacks spatial coverage but spans long periods of time (approximately 2 months). There is a need to understand the extent over which results from each survey type might be applied—that is, how much time is represented by the results from a single mobile survey, and how much space is represented by the results from the stationary surveys.
The goal of this project was to use each of these two complementary methods to inform our understanding of the results from the other. Specifically, the mobile acoustic survey data were used to provide an estimate the spatial representative range of the stationary results. The stationary data were be used to estimate the temporal representative range of the 24- hour mobile survey results. Concurrent data collected by the two methods were also compared to assess the challenges associated with each survey type, and to confirm whether both methods provide similar findings.
This assessment utilized backscatter data from repeated passes of one of the mobile transects, and from 3 of the two-month deployments of the stationary platform. The spatial representative range of the stationary results could not be determined using data from the single transect. However, the stationary dataset revealed strong tidal and diel periodicities in volume backscatter (roughly proportional to fish density) at this site, with greater variation occurring at these small time scales than over course of the year. This finding reinforces the importance of 24-hr data collection periods in ongoing monitoring efforts. Collecting at least 24 hours of data at a time allows this tidal and diel variability to be quantified and kept separate from the longer-term trends that we seek to monitor. The temporal representative range of a 24-hr survey was determined to be approximately 3 days. At the 24-hour scale, water column backscatter was comparable across the two survey types, but at shorter time scales, it was not.
Data from both survey types were subject to contamination by backscatter from entrained air in the water column—a common issue at tidal energy sites. All data had to be carefully scrutinized and cleaned, which was an extremely time-consuming process and highlights the need to develop more advanced backscatter classification tools.