Vol. 22: 125–143, 2013 ENDANGERED SPECIES RESEARCH Published online December 2 doi: 10.3354/esr00538 Endang Species Res OPENPEN ACCESSCCESS Using drifting passive echolocation loggers to study harbour porpoises in tidal-stream habitats Ben Wilson*, Steven Benjamins, Jim Elliott Scottish Association for Marine Science, Scottish Marine Institute, Oban, Argyll and Bute PA37 1QA, UK ABSTRACT: Moored passive acoustic detectors (e.g. C-PODs) are widely used to study harbour porpoise Phocoena phocoena distribution and relative abundance, but their use in tidal-stream habitats is complicated by the need for retrievable flow-resistant seabed fixings and the occur- rence of flow-induced noise in the resultant data. In this study, we explored the use of a new method aimed at tidal-stream habitats, which are of increasing interest for marine renewable energy generation. Porpoise detectors (C-PODs) were attached to multiple drifters and repeatedly set adrift at a tidal-stream site in western Scotland during May 2010 and August 2011. Porpoise vocalisations were successfully detected under varying tidal conditions during approximately 63 h of drifting. Harbour porpoise distribution, as determined by the drifting detectors, was similar to that found using the traditional, yet more logistically intensive, visual and acoustic boat-based sur- veys and to an extent that found by moored C-PODs. Drifting detectors also mapped tidally driven spatiotemporal variability in ambient noise levels which could influence porpoise detection. In summary, drifters equipped with passive acoustic detectors offer a new, rapid and inexpensive tool for investigating porpoise occurrence and behaviour in tidal-stream habitats, and should be considered as part of a comprehensive marine mammal monitoring approach of these energetic environments in the context of marine renewable energy development and other industries. KEY WORDS: Phocoena phocoena · Tidal stream · Marine renewable energy · Survey · Methodology · Drifter INTRODUCTION Inger et al. 2009, ICES WGMME 2011, Frid et al. 2012). Cetaceans are thought to be at risk from tidal There is global and growing interest in exploiting energy devices in various ways, including collision marine renewable energy resources, including those with devices, disturbance during construction and/or associated with tidally driven water movement, par- device operation, noise emissions and habitat exclu- ticularly through narrow channels, around head- sion (Carter et al. 2008, ICES WGMME 2011). In lands or other areas of high current flow (sometimes order to address — and perhaps mitigate — these exceeding 4 m s−1; WEC 2010). Numerous devices potential issues, we need to know how and why ceta - capable of extracting this energy are currently under ceans are already using areas of interest for energy development (e.g. EMEC 2012). In Scotland, there is extraction, particularly in terms of small-scale distri- considerable interest in further developing this in - bution and habitat use. Currently, however, these dustry, given Scottish Government policy ob jectives issues are poorly understood, largely due to the diffi- of meeting 100% of gross annual electricity demand culties of studying these highly mobile species in from renewables by 2020 (SG 2011). such energetic environments using standard tech- Concerns have been raised over potential effects of niques. Boat-based line-transect surveys, for exam- energy extraction on tidal-stream environments (e.g. ple, are a standard method of assessing cetacean © The Authors 2013. Open Access under Creative Commons by *Email: [email protected] Attribution Licence. Use, distribution and reproduction are un - restricted. Authors and original publication must be credited. Publisher: Inter-Research · www.int-res.com 126 Endang Species Res 22: 125–143, 2013 abundance, but animals are often difficult to see at increasing the risk of damage or loss, and/or interfere the surface in tidal streams because of turbulence, with their recording sensitivity. Tidal streams also standing waves and other intermittent features cor- produce elevated ambient sound levels which can related with tidal flow. Furthermore, basic assump- mask cetacean sounds, particularly during peak tidal tions of the underlying distance sampling methodol- flow; finally, the rapid flow of water past the hydro - ogy may be violated, particularly in narrow channels phones in the detectors adds self-noise to the data between islands where currents may preclude an (Au & Hastings 2008, Bassett et al. 2010). unbiased distribution of survey effort, edge effects Because of these technical and methodological may be significant, and water movement may cause challenges, comparatively little is known about small- animals to be non-randomly distributed relative to scale spatial distribution and habitat use of cetaceans survey transects (Buckland et al. 2001, 2004). Given in tidal streams. Nonetheless, with in creasing num- that the water containing the animals may be moving bers of tidal energy sites being considered for devel- at an appreciable fraction of the speed of the survey opment in Europe, North America, New Zealand and vessel (potentially violating a central assumption of elsewhere, studying cetaceans in these habitats is distance sampling that animals are stationary), counts becoming more urgent, to: and resulting density estimates may be severely (1) identify pre-development baseline levels of distri- biased depending on the direction of travel of both bution and relative abundance across different animals and the survey vessel relative to the current. temporal scales (tidal, diel, seasonal); These problems are less pronounced in aerial sur- (2) determine which features of the tidal stream envi- veys because water speeds are insignificant com- ronment are especially attractive to cetaceans, pared to the aircraft’s own speed. However, as most and why; and tidal sites are small (<1 to 10 km wide) and spatially (3) inform what impacts might occur following con- heterogeneous, these high speeds will negatively struction and during long-term device operation. bias sighting rates, exacerbated by poor sighting con- Despite the problems outlined above, there are ditions at peak flow due to tidal turbulence. Any sur- good reasons to consider using passive acoustic mon- vey programme would also need to consider variabil- itoring approaches, including their ability to record ity across tidal cycles, requiring significant effort. under conditions unsuitable for visual observations. Visual observations from adjacent coastal vantage Several passive acoustic click detectors/loggers points can be informative but will be limited by the are currently available, including the C-POD (Chelo- availability of suitable sites and thus biased towards nia Ltd. see www. chelonia. co. uk/ about_ the_cpod . htm) particular locations and habitat features (Pierpoint and PAMBuoy® (Marine Instrumentation Ltd. see 2008). www. wildlifeacoustics. com/ products/song-meter-sm2- Acoustic monitoring faces comparable problems to plus-submersible). These devices record evidence of shipboard visual observations or fixed vantage point vocalising porpoises and other small cetaceans by approaches when applied in tidal streams. Such either recording actual sounds or by logging a record monitoring is typically carried out using either an of each detected event. Of these devices, C-PODs are array of hydrophone elements towed behind a survey currently in widespread usage worldwide. C-PODs vessel (offering good spatial coverage; e.g. Barlow & use waveform characteristics to identify odontocete Taylor 2005, Barlow 2006, Lewis et al. 2007) or by echolocation clicks among other broadband sound using moored autonomous acoustic recorders (offer- sources. They log time, duration, centre frequency, ing good temporal coverage; e.g. Nieukirk et al. loudness, inter-click interval and bandwidth of each 2004, Simard et al. 2008, Van Parijs et al. 2009). received click. Although originally developed to Strong directional currents in tidal streams make it detect high-frequency narrow-bandwidth clicks pro- difficult to maintain hydrophone arrays in the correct duced by species such as harbour porpoises Pho- configuration and generally to safely manoeuvre the coena phocoena, C-PODs can also identify lower- survey vessel. Fixed autonomous acoustic recorders frequency/broad-band clicks produced by other require robust moorings to resist the current, adding odontocetes such as bottlenose dolphins Tursiops to mooring weight, complexity and cost, and poten- truncatus. Other para meters recorded include device tially requiring larger vessels to safely deploy and tilt from vertical, ambient temperature and a crude retrieve them (Dudzinski et al. 2011). Furthermore, proxy for background noise levels. C-PODs (and T- mooring deployment and retrieval may only be possi- PODs, their predecessors) have sufficient battery and ble during brief periods of slack water. Strong cur- memory capacity to remain de ployed for up to 3 mo. rents may deflect recorders towards the substrate, These devices are widely used to investigate tempo- Wilson et al: Studying porpoises using drifting acoustic detectors 127 ral patterns in odontocete occurrence and distribu- passing over the detector, resulting in cleaner tion to study basic ecology as well as potential recordings (Au & Hastings 2008). Third, detectors anthropogenic impacts of marine industries (Cox et sample a greater area than when moored. Finally, al. 2001, Carlström