Spiddal - Ocean Energy Test Site

Home , Aran Islands, Galway, Galway Bay, Spiddal

CETACEAN PRESENCE AT THE OCEAN ENERGY TEST SITE

SPIDDAL: AS DETERMINED THROUGH LAND-BASED VISUAL

MONITORING AND STATIC ACOUSTIC MONITORING USING

PODS

Report prepared for the Marine Institute

Dr. Joanne O’Brien

Galway-Mayo Institute of Technology

March 2013

Ç /

1 Introduction ...... 1

2 Methods ...... 3

2.1 Study Area ...... 3 2.2 VISUAL SURVEYING TECHNIQUES ...... 4 2.3 Static Acoustic Monitoring (SAM) Equipment ...... 5 2.3.1 T-PODs ...... 5 2.3.2 C-PODs ...... 6 2.4 SAM data analyses ...... 9 2.5 Moorings used in Galway Bay ...... 11

3 Results ...... 12

3.1 Visual Monitoring...... 12 3.2 SAM using T-PODs (SAM 1) ...... 13 3.3 SAM using C-PODs (SAM 2) ...... 16 3.4 Assessment of site usage ...... 18 3.5 An assessment of potential effect of an OE device on harbour porpoise presence ...... 19

4 Discussion ...... 21

5 References ...... 23

1 INTRODUCTION

Historical reviews of sightings, strandings and captures of cetaceans in Ireland were published by Schariff (1900), Moffat (1938), O’Riordan (1972) and Berrow and Rogan (1997) among others (Table 1), which include a number of references to cetaceans in Galway Bay. The only contemporary data on the distribution and abundance of cetaceans in Galway Bay is from sightings data collected as part of the Irish Whale and Dolphin Group (IWDG) sighting scheme in operation since 1991and more recently through dedicated work carried out by O’Brien (2009) and Berrow et al., (2008). Of the 24 cetacean species recorded in Irish waters, 16 species have been recorded in Galway Bay, seven species have been recorded both stranded and visually observed, and two species have only been recorded observed, while nine species are known only to occur through strandings (Table 1). There are limitations associated with strandings data, as it may be that the animals washed up originated outside of the study area, and therefore provide false data on the presence of certain species in an area. Berrow et al. (2002), analysed 2,200 visual sighting records from the Irish Whale and Dolphin Group (IWDG) and found 13.2% of all records were from Co. Galway, with harbour porpoises been the most frequently reported species, and Galway producing the third most sightings of the species in the country. Most records were reported between June and August, with few sightings in the winter and spring. Berrow et al. (2002) also showed that Bottlenose dolphins were the third most frequently sighted species in the country, with concentrations of sightings occurring within Galway Bay. They also showed that bottlenose dolphin sightings increased rapidly from April to June, suggesting an inshore movement, which peaked in August. In more recent years this was not found to be the case. O’Brien (2009) found that Harbour porpoise was the most regularly recorded species but dolphin sightings of any species were rare. Targeted research for abundance estimation was carried out in Galway Bay in 2008 (Berrow et al., 2008) and results showed an overall density estimate of 0.73 per km2 (CV=0.21) with an abundance ±SE of 402±84 and 95% Confidence Intervals of 267-605 animals in the bay. The proportion of juveniles and calves in the bay at this time was calculate at 7% and was consistent with other areas also surveyed at this time (Berrow et al., 2008). Density estimates for harbour porpoise in Galway Bay are similar to those at the Blasket Island, one of Ireland candidate Special Areas of Conservation (cSAC) for the species Berrow et al. (2009).

Table 1: Cetacean species recorded either visually or through the recording of strandings in Galway Bay.

Species Visually Reference Stranded Reference recorded

Bottlenose dolphin * Cooke (1990) * Moffat (1938) Harbour porpoise * www.iwdg.ie * O’Riordan (1976) Common dolphin * www.iwdg.ie * O’Riordan (1972) Killer whale * McGrath (1983) * www.iwdg.ie Minke whale * www.iwdg.ie * Fairley (1998) Pilot whale * www.iwdg.ieF * Fairley (1979) Risso’s dolphin * www.iwdg.ieF * D’arcy Thompson (1900) Sperm whale * www.iwdg.ieF * Cabot (1967) False killer whale * O’Cadhla et al., 2004 Atlantic white- * Fairley and Dawson (1981) sided dolphin Cuvier’s beaked * Andersen (1904) whale Fin whale * Harmer (1914-27)

Humpback whale * Berrow et al., 2006 Northern * Fraser (1934) bottlenose whale Pygmy sperm * Fairley and Mooney (1985) whale Sowerbys beaked * Harmer (1914-27) whale Striped dolphin * Fairley & MacLoughlin

(1990) Trues beaked * Harmer (1914-27) whale

Visual monitoring carries many constraints and is influenced by variables such as sea state (Evans and Hammond, 2004; Teilmann, 2003; Palka, 1996; Clarke, 1982), observer variability (Young and Peace, 1999; O’Brien et al, 2006), optics and height above sea level. Evans and Hammond (2004) state that visual surveys should generally not be carried out in sea states above Beaufort scale 2, as the probability of detecting animals is markedly reduced above this. Therefore on the West coast of Ireland Static Acoustic Monitoring is especially useful for monitoring small vocal cetaceans since it can be carried out without the interference of the variables mentioned above, and, most importantly, does not negatively impact upon the animals. SAM involves the detection and recording of cetacean vocalisations or echolocation clicks and is a very valuable tool for the exploration of fine scale habitat use by the various

¡ odontocete species. Additionally SAM can be used to effectively assess habitat use of cetacean species and is particularly useful for the study of behaviour, such as feeding strategies, approach behaviour and communication. Significant effects of diel pattern have been described in the foraging behaviour of harbour porpoise (Carlström, 2005; Todd et al, 2009). SAM will not provide robust information on density and abundance of cetaceans in an area, but gives valuable information on spatial and temporal trends. This report was prepared for the Marine Institute and gives a detailed account of cetacean occurrence at the OE Test site, east of Spiddal Co. Galway on the north shore of Galway Bay in preparation for their foreshore licence application to install a standard telecommunications cable from the new pier at Spiddal west to the ocean energy test site. Archived visual and acoustic datasets held at the Galway-Mayo Institute of Technology (GMIT) from the years 2005-2007 were accessed as well as data from the PReCAST project 2009-2010 (O’Brien et al., 2012) to meet requirements from the Development Applications Unit in the Department of the Environment, Community and Local SAM datasets from the OE Test site are the longest recorded in Ireland with c900 days monitored across all studies. This report sets out how all data were collected and explores the distribution and occurrence of small cetaceans, specifically harbour porpoise Phocoena phocoena at the OE site. Such information will inform managers when are suitable times to target works in the area to ensure disturbance is mitigated against and kept to a minimum.

2 METHODS

2.1 Study Area

Galway Bay is situated on the west coast of Ireland and is bounded by the northern and southern shores of counties Clare and Galway and lies between the lines of longitude of 8º55’W and 9º50’W and latitude of 53º00’W and 53º15’N (De Bhaldraithe 1977), (Figure 1). It is one of the largest bays on the west coast of Ireland, and is about 50 kilometers long and from 10 to 30 kilometers in breadth. A chain of three islands, the Aran Islands, stretches across the mouth of the bay. These form a partial boundary between the bay and the Atlantic Ocean. Water exchange between the bay and the Atlantic is between four sounds, the North and South Sounds, Gregory and Foul Sounds. The meridian of 9º16’W between Black Head and Spiddal conveniently divides Galway Bay into the inner and outer bays (Lei 1995). Depths range between 8-20m in the inner bay and 20-60m in the outer bay (Nolan 1997; Lei 1995). Tidal range during springs is 4.5m and during neaps is 1.9m. The main freshwater influence in the bay comes from the River Corrib, while the Clarinbridge and Kilcolgan rivers also have a

¢ marginal contribution. This freshwater influence is restricted mainly to the north shore (Fernandes 1988). The Galway Bay Complex SAC (000266) comprises a diverse range of marine, coastal and terrestrial habitats and includes some of the best examples of shallow bays, reefs, lagoons and salt marshes in the country (Galway Bay Complex, Site Synopsis, www.npws.ie). The site supports an important common seal colony and a breeding otter population, both of which are listed under Annex II of the EU Habitats Directive but no cetacean species are listed as qualifying interests of the site.

Figure 1. Map of Galway Bay study area, including site where C-PODs and T-PODs were deployed.

2.2 VISUAL SURVEYING TECHNIQUES

Monthly dedicated land-based visual surveying was carried out from Spiddal Pier to monitor the seasonal occurrence, distribution and abundance of cetacean species in the vicinity, between March 2005 and May 2007. The same single observer was used during all observations to reduce any inter-observer variability. Optical equipment used throughout land-based quantified effort watches included Opticron 7x50 binoculars and Kowa TGW2 with 20X wide eyepiece. All visual observations were carried out in Beaufort Sea states 2 or less following recommendations by Evans and Hammond (2004). All watches were 100 minutes in duration, enabling the amount of effort to be quantified, and therefore allowing the generation of relative abundance estimates (animals sighted per hour). Environmental conditions (sea

£ state, wind speed and direction, cloud cover and visibility) were recorded for the duration of all surveys. When a sighting was made, the species observed was identified, as well as the numbers of groups and individuals present. A group was defined after Shane (1990) and Smolker et al. (1993), as a solitary animal or aggregation of animals, observed in apparent association, moving in the same direction, and exhibiting similar behaviour where a member was within 10m of any other member in the first 10 minutes the animals were observed. The presence and numbers of juveniles or calves were also noted. A calf was defined as two-thirds or less the length of an adult (Shane 1990). The distance of the observed animals from land was estimated and the behaviour type recorded. Further details noted included direction of travel and surfacing mode.

2.3 Static Acoustic Monitoring (SAM) Equipment

2.3.1 T-PODs The T-POD is equipped with a hydrophone element which is connected to two band pass filters, a comparator/detector circuit and a microprocessor which has memory capability to store information logged from target cetacean species (Kyhn, 2006). All electronics are contained within a waterproof PVC housing (Figure 2). The dedicated software T-POD.exe is used to set and download the data from the logger, which can identify and classify click trains of cetacean origin. Once deployed, a T-POD runs six successive scans each of 9.3 seconds duration, and selects only tonal clicks and logs the time and duration of each click. However, sensitivities between units differ and tank calibration tests are recommended prior to their deployment. These tests should determine the detection threshold of each unit as this is directly related to detection range (Kyhn et al., 2008). In addition, field calibrations are also recommended prior to employment of the devices in monitoring programmes in order to facilitate comparisons between datasets collected in different areas using multiple loggers (Dähne et al, 2006). A detection distance of over 1,000m for T-PODs and bottlenose dolphins was generated in the Shannon Estuary by Philpott et al., (2007) using version three T-PODs, but it is likely that this may differ with more recent versions. Detection distances for the harbour porpoise using T-PODs were generated by Tougaard et al., (2006) (200m) and Villadsgaard et al., (2007) (300m to 500m). The T-POD.exe software is required for setting and downloading T-PODs, either through the use of a printer port cable or more recently, USB. The T-POD trains used to process .pdc files for trains and the filter is based on an algorithm that uses a 38% increase or decrease in an interval as the constraint. The true value for small odontocete trains is occasionally much higher but cannot be implemented in practice without very complex processing and/or a high level of false positive trains (Chelonia Ltd).

¤ Train selection is categorised by the probability of a train being of cetacean origin. Data can be exported under various parameters and displayed on text or .csv filesField calibrations of T- PODs were not carried out prior to the commencement of these early studies and the settings used during deployments are presented in Table 2. The first SAM carried out in Galway Bay used version 4 and 5 T-PODs. However these units are no longer in production but are still used to monitor cetaceans in the wild.

Table 2. Generic settings for T-PODs as recommended by Chelonia Ltd

T-POD generic settings SCAN 1 2 3 4 5 6 A filter (kHz) 50 130 50 130 50 130 B filter (kHz) 70 92 70 92 70 92 Click bandwidth 5 4 5 4 5 4 Noise adaptation ++ ++ ++ ++ ++ ++ Sensitivity 6 6 6 6 6 6 Scan limit 240 240 240 240 240 240

Figure 2. T-POD, version 5 unit by Chelonia Ltd

2.3.2 C-PODs

Once deployed at sea, the C-POD operates in a passive mode and is constantly listening for tonal clicks within a frequency range of 20 to 160 kHz (Figure 3). When a tonal click is detected, the C-POD records the time of occurrence, centre frequency, intensity, duration, bandwidth and frequency of the click (Chelonia Ltd). Internally, the C-POD is equipped with a Secure Digital (SD) flash card, and all data are stored on this card. Dedicated software, CPOD.exe, provided by the manufacturer, is used to process the data from the SD card when

¥ connected to a PC via a card-reader. This allows for the extraction of data files under pre- determined parameters, as set by the user. Additionally, the C-POD also records temperature over its deployment duration. It must be noted that the C-POD does not record actual sound files, only information about the tonal clicks it detects. The C-POD detector is a sound pressure level detector with a threshold of 1Pa peak to peak at 130 kHz, with the frequency response shown below (Figure 3).

Figure 3. Threshold for detection across various frequency bands between 20 and 200 kHz for the C-POD (note 1Pa p-p is the SI unit for pressure and correctly represents the threshold) © Chelonia Ltd

Calibration of equipment is important in order to compare results across units. Chelonia Ltd calibrates all units to a standard prior to dispatch. These calibrations are carried out in the lab under controlled conditions and thus Chelonia highly recommends that further calibrations are carried out in the field prior to their employment in monitoring programmes instead of further tank tests (Nick Tregenza pers comms). All C-PODs deployed in Galway were calibrated during field trials prior to deployment in a long-term monitoring programme. A detection distance of 797.6m ±61m (75% of groups recorded<400m) for C-PODs and bottlenose dolphins was generated in the Shannon Estuary while distances of 441m ±42m (92% <400m) were generated for the Harbour porpoise in Galway Bay (O’Brien et al., (Submitted)).

¦ Screw top end and safety line Hydrophone attached to element middle

Figure 4. C-POD unit by Chelonia Ltd

Through the C-POD.exe software, data can be viewed, analysed and exported. Additionally, the software can be used to change settings of individual SD cards. The software includes automatic click train detection, which is continually evolving as Chelonia Ltd receives more feedback from their clients. The C-POD.exe software (example Figure 5) is very similar to the T-POD.exe but has capabilities beyond its predecessor. C-POD.exe can be run on any version of Windows and requires an external USB card reader, which reads the SD card into the directory. Version 2.013 was used for all analyses. C-POD.exe software allows the user to extract click trains under five classification parameters:

i) porpoise-like ii) dolphins iii) other train sources iv) unclassed v) boat sonars.

Figure 5. Screen grab of C-POD.exe, showing a harbour porpoise click train

2.4 SAM data analyses

All T-POD and C-POD data were analysed similarly to allow for comparison between the two studies. Only high probability clicks were used during all analyses. Both dolphin and porpoise detections were extracted as detection positive minutes per day and per hour hourly extractions allowed for classification into the following categories; season (spring, summer, autumn and winter), diel cycle (day and night-time), tidal state (ebb, flood, slack high, slack low) and tidal phase (spring, neap). The term PPM represents the number of minutes in a day or an hour that harbour porpoises were acoustically detected, while DPM represent the number of minutes dolphins were detected. The term encounter refers to the detection of a series of clicks/click trains followed by a period of quietness at least 10 minutes in duration. Seasonal categorisations were assigned according to the seasons spring (February, March, April), summer (May, June, July), autumn (August, September, October) and winter (November, December, January). Data files in the format PPM/h and DPM/h were divided into day and night-time categories using local times of sunrise and sunset times, obtained from the U.S. Naval Observatory (www.aa.usno.navy.mil/data/docs/RS). Hourly data segments were further categorised into each of the four tidal states, where three hours was assigned to each state (one hour either side of the hour). Files were further split to correspond with tidal phase (spring and neap cycles) using admiralty data (WXTide 32) where two days either side of the

¨ highest tidal height was deemed spring, and two days either side of the lowest tidal height was deemed neap. All data were statistically analysed using the programme R. A GLMM was fitted to the binomial data using the glmer function in the lme4 package developed for R. C-POD/T-POD ID number was included as a random factor to take into account variability between units. Akaike’s information criterion (AIC) and a histogram of fitted residuals were used as diagnostic tools for model selection. Wald chi-squared tests were computed for each variable and predicted proportions of DPH were extracted across all levels and displayed as box plots using the HH package developed for R. R is a language and environment for statistical computing and graphics. It is free software, available at http://www.r-project.org/index.html. The software compiles and runs on a wide range of UNIX platforms, Windows and MacOS. R provides a wide variety of linear and nonlinear modelling, classical statistical tests, time-series analysis, classification, clustering and graphical techniques (R Development Core Team, 2011). R is designed around a true computer language, similar to the S language (see Appendix for full R scripts used). The effective programming language includes conditionals, loops, user-defined recursive functions and input and output facilities. Further analyses were carried out on the C-POD data as SAM datasets can be explored to assess cetacean behaviour which can provide an important insight into how small cetaceans use a site. With the advanced nature of the C-POD.exe software, accurate details on the frequency and inter-click interval (ICI) of individual cetacean click trains can be investigated. Details of individual harbour porpoise click trains were extracted and analysed. Trains with a minimum ICI of 10ms were selected as feeding buzzes and all other trains categorised as other. All trains were categorised according to diel, tidal cycle, phase and season and further statistical tests applied to assess id the deployment location is an important feeding site. All statistical analyses of the data were carried out using the programme R. A generalized linear mixed effect model (GLMM) was fitted to the binomial data, using the glmer function in the lme4 package developed for R where MinICI<10ms = 1, termed “feeding buzzes” (foraging) and >10ms = 0 (not foraging). Akaike’s information criterion (AIC) and a histogram of fitted residuals were used as diagnostic tools for model selection. C-POD ID was included in the GLMM model as a random factor to take into account intra POD variability over the project duration. Wald chi-squared tests were computed for each variable and predicted proportions of MinICI<10ms were extracted across all levels and displayed as box plots using the HH package developed for R.

Two mooring types were used during the Galway Bay SAM studies, light weight moorings (Figure 6) but deployments were also carried out from the wave energy device itself at Spiddal (Figure 7).

SAM unit

Figure 6. Light weight moorings as erected at the OE test site in Galway Bay

Figure 7. Mooring design used for deployment of units from the Smart Bay Buoy network

3 RESULTS

3.1 Visual Monitoring

A total of 28 dedicated land-based visual watches were carried out from Spiddal Pier (2700 minutes/45 hours) between March 2005 and February 2007. Cetaceans were recorded during 10 of the 27 watches (37%). A total of 16 sighting were recorded during watches comprising of three species; including, harbour porpoise (81%), bottlenose dolphin (13%) and Minke whale (7%). Only a single sighing of two harbour seals was recorded on one occasion. All sightings recorded were within a 5km radius of Spiddal pier. Most sightings (75%) were recorded between the months June to December with only 25% of sightings recorded in the period January to May, highlighting mid-summer through to December as the months when porpoises are most active at the site. These results are similar to results from SAM 1 and SAM 2 (see below for SAM results). However, when data is compared to other sites in the bay, Spiddal isn’t the most important (Table 3).

Figure 8. Visual sighing data as recorded from Spiddal Pier, March 2005 to February 2007.

Table 3. Calculations of harbour porpoise relative abundance from Black Head, Spiddal and Fanore, 2005-2007.

Site No.of No. of % Total no. of HP Relative abundance watches watches when (harbour porpoises HP recorded hour-1) Black Head 31 18 58 110 2.12 hr-1 Spiddal 28 8 29 32.5 0.69 hr-1 Fanore 29 10 36 38.5 0.79 hr-1

3.2 SAM using T-PODs (SAM 1)

T-PODs were first deployed in Galway Bay on the 12 of May, 2006, and were subsequently deployed for various periods thereafter, when weather permitted retrieval and re-deployments (Table 4) until October 2007. A total of 333 days were monitored at the site.

Table 4. Details of deployment locations and T-POD numbers assigned to the site over the duration of the study (SAM 1).

Location Site Deployment Recovery T-POD Deployment date date No. duration Galway Bay Spiddal 12.05.2006 17.06.2006 404 36d 8h 0m Galway Bay Spiddal 04.07.2006 23.12.2006 505 43d 21h 58m Galway Bay Spiddal 03.10.2006 09.11.2006 451 37d 4h 37m Galway Bay Spiddal 09.11.2006 23.12.2006 324 43d 23h 41m Galway Bay Spiddal 01.02.2007 26.03.2007 505 52d 23h 18m Galway Bay Spiddal 26.03.2007 12.04.2007 506 0d 0h 19m Galway Bay Spiddal 12.04.2007 12.06.2007 652 61d 9h 38m Galway Bay Spiddal 12.06.2007 10.07.2007 568 28d 0h 03m Galway Bay Spiddal 10.07.2007 01.08.2007 506 21d 18h 41m

Results showed that harbour porpoises were detected on average during 88% of days monitored at Spiddal, while dolphin detections were only recorded 3% of the time (Table 3). The highest number of harbour porpoise detection positive (PPM) minutes (1001dpm) was recorded from Spiddal in October, 2006 (Table 5/Figure 9).

Table 5. Detection details from T-POD deployments at Spiddal

Details Porpoise Dolphin detections detections Year Location Month No. days % of days Total % of days Total deployed with PPM with DPM porpoise dolphin detections detections 2006 Spiddal May 20 100 241 5 1 Spiddal June 17 94 165 0 0 Spiddal July 28 100 271 4 1 Spiddal August 17 94 129 6 1 Spiddal October 29 97 1001 0 0 Spiddal November 31 100 637 5 1 Spiddal December 23 100 265 0 0 2007 Spiddal February 28 43 22 0 0 Spiddal March 26 58 50 4 1 Spiddal April 19 100 179 0 0 Spiddal May 31 97 373 0 0 Spiddal June 19 100 136 0 0 Spiddal July 10 100 102 0 23 Spiddal September 30 73 104 7 2 Spiddal October 5 80 16 0 0

Figure 9. Harbour porpoise detections per month

Data from both years monitored with T-PODs were pooled for statistical tests due to lack of data for some months in some years. This was due to T-POD malfunction as they were less reliable than current digital models (C-PODs). The T-POD dataset across years showed season to have a significant effect on detections with significantly less detections in Spring compared to all other months (2=190.3, p<0.0001). This is similar to results produced through visual monitoring. Peaks in detections occurred during the autumn and winter. However, no significant variation in detections was present across diel and tidal cycle (Figure 10).

Figure 10. Predicted proportion of detection positive hours, in the porpoise channel (92 – 130 kHz) at Spiddal (Galway Bay) across the three variables of season; diel, where D =day and N = night; and tidal cycle, where E =ebb, L = slack low, F= flood and H=slack high.

3.3 SAM using C-PODs (SAM 2)

The OE Test site was again the target of long-term SAM as part of the PReCAST project, when monitoring commence in January 2009 and continued until September 2010. During this time a total of 572 days were monitored at the site. Results show that, on average, harbour porpoises were recorded on 95% of days monitored, while dolphins were rarely recorded (4% days). These results reflect those of the SAM 1. Over the 572 days monitored, a total of 27,902 porpoise Detection Positive Minutes (DPM) were recorded (4,515 Detection Positive Hours; DPH). As dolphin sightings were rare, only the porpoise data were analysed to identify factors influencing their presence at the site. Results are presented for 2009 and 2010 separately due to the robustness of the dataset. In the 2009 dataset, season was shown to significantly affect porpoise presence at the site (2=58.8, p<0.0001), where a peak in harbour porpoise occurrence was observed through autumn and winter. Results from the model also highlight diel cycle to contain significant variation (2= 26.7, p<0.0001), indicating that night and morning phases have a higher level of harbour porpoise detection rates. A significant variation

across tidal phase (2= 36.1, p<0.0001), shown in Figure 11, exists between neap tide and spring tide with a rise in DPH during neap tide. Results suggest the significance of tidal cycle (2=39.6, p<0.0001), which can be most likely attributed to the predicted drop in detections during slack low tide.

Figure 11. Predicted proportion of detection positive hours, in the narrow band high frequency channel at Spiddal (Galway Bay) 2009 across the four variables of season; diel, where D =day, E= evening, M= morning and N = night; tidal phase, where Trans.=transitional phase, NT= neap tide and ST=spring tide; and tidal cycle, where E =ebb, L = slack low, F= flood and H=slack high

Results from 2010 dataset indicate a change in seasonal pattern (2=113.8, p<0.0001), where a peak in harbour porpoise occurrence was observed during winter but, in contrast to results from 2009, autumn contained fewest NBHF detections. This was due to the absence of data for the month of October due to device malfunction. Results highlight diel cycle to contain significant variation, although in reference to Figure 11, a distinct diel pattern is unclear and the comparatively low chi-squared value derived for this variable reflects this (2=25.1, p<0.0001). A significant variation across tidal phase (2= 16.7, p=0.0008) concurs with 2009 findings, with a rise in detections during neap tide. Results suggest a significance of tidal cycle (2=23.1, p=0.0001), with a slightly higher level of detections during an ebbing tide (Figure 12).

Figure 12: Predicted proportion of detection positive hours, in the narrow band high frequency channel at Spiddal (Galway Bay) 2010 across the four variables of season; diel, where D =day, E= evening, M= morning and N = night; tidal phase, where Trans.=transitional phase, NT= neap tide and ST=spring tide; and tidal cycle, where E =ebb, L = slack low, F= flood and H=slack high.

3.4 Assessment of site usage

Graphs of MinICI were generated for NBHF trains detected at Spiddal from C-POD data only. A total of 144,216 NBHF click trains were recorded at Spiddal using C-PODs over the deployment period. The average number of clicks per train was 15, with on average 175.5 clicks recorded per second, and with an average frequency of 130.7 kHz across all deployments. Click trains were classified into two categories based on the data presented above, where the category foraging was applied to trains with MinICI<10ms. All other trains were defined as “Other” as no definite behaviour category could be attributed. Results showed 41% (60,386 trains) of the total click trains recorded fell under the category foraging, highlighting the site at Spiddal as a regular feeding spot. Season was found to be the most significant variable (2 = 3282.4, P>0.001), with the highest levels of feeding buzzes during winter. Within tidal cycle (2 = 100.4, P>0.001), the highest level of feeding buzzes was found

during an ebbing tide (Figure 13), and results also show that the diel category “night” contains the highest predicted proportion of feeding buzzes (2 = 1053.4, P>0.001). Tidal phase was found to be the least significant predictor of feeding buzzes (2 = 13.9, P>0.001), and the low chi-squared value indicates that this variable may only be highlighted due to the large dataset.

Figure 13: Predicted proportion of NBHF (narrow band high frequency) click trains with minimum inter-click intervals of less than 10ms (MinICI<10ms) in Galway Bay across the four variables of season; diel, where D =day, E= evening, M= morning and N = night; tidal phase, where Trans.=transitional phase, NT= neap tide and ST=spring tide; and tidal cycle, where E =ebb, L = slack low, F= flood and H=slack high.

3.5 An assessment of potential effect of an OE device on harbour porpoise presence

As harbour porpoises were detected frequently at the OE test site off Spiddal during the SAM 2 study, it afforded an opportunity to test its potential effect on the presence of cetaceans (O’Brien et al., 2012). SAM was carried out on a continuous basis at the wave platform, so it was decided to assess if there was a difference in detection rate at two additional sites, 1,000m east of the device and 500m west of the device. Light weight moorings were established at each of these additional sites and a single C-POD was deployed. The presence of the wave

platform, which is of substantial size (28 tonne), could have had a positive or negative effect on the occurrence of harbour porpoises in the area:

• The presence of such a structure may deter animals. They may not be able to sufficiently forage for food as the structure may impact on their echolocation ability. This event is highly unlikely at Spiddal given the high percentage of days with detections.

• Or the platform itself may act as a cover for many fish species and, therefore, attract fish to the area and, in turn, feeding porpoises. International studies have found that wave buoys can serve as artificial reefs and attract fish and other marine life. In fact, in some parts of the world, conventional buoys are deployed to serve as "Fish Attracting Devices" (FADs) (Nelson, 2003).

Results from this short deployment failed to show a significant difference in detections between sites (P=0.001), suggesting the OE platform did not influence harbour porpoise presence (Figure 14).

Figure 14: Results from C-POD deployments from LW Ms and ES-W P

! "

4 DISCUSSION

Cetaceans live in an acoustic world and increasingly attempts have been made to develop acoustic monitoring techniques rather than relying on visual methods, whose efficiency is hugely dependent on light, weather conditions and sea-state, especially for species such as the elusive harbour porpoise. Additionally, the reliance on sound by these animals is extremely important and therefore SAM is a very valuable tool for determining presence and assessing fine scale habitat use by various odontocete species. The main advantage of SAM is that it can provide information on species that can go undetected visually for up to 87.1% of the time (bottlenose dolphin Tursiops truncatus Montagu; Mate et al., 1995) and 95% (harbour porpoise; Read & Westgate, 1995). Patterns of cetacean presence have been described over seasonal scales (Canning et al., 2008, Bolt et al., 2009; Simon et al., 2010; Gilles et al., 2011) diel cycle (Cox & Read 2004; Carlström, 2005; Todd et al., 2009; Phillpott et al., 2007) and tidal patterns (Philpott et al., 2007; Marubini et al., 2009). In order to evaluate the importance of an area, it is fundamental that the presence of small cetaceans at a site is fully understood and this requires monitoring over varying time scales depending on monitoring methods. When the visual and SAM datasets from Spiddal are compared, it is evident that the SAM dataset is much more complex and gives a better reflection of cetacean activity at the site, but it fails to inform on the numbers present. The aim of the present study was to compile archived data from available sources and to explore the presence of small cetaceans in the vicinity of the OE Test Site, east of Spiddal on the north shore of Galway Bay. These results were taken from O’Brien (2009) and from the PReCAST project designed to address a wide range of issues and will contribute to developing policy advice on meeting Irelands’ statutory obligations (O’Brien et al., (2012). Results from visual and acoustic monitoring are very similar as all show that autumn and winter months are when porpoises are most active at the site. Visual data shows that in comparison with other sites in the bay subjected to land-based watches, Spiddal is not the most important, with a greater relative abundance recorded from Black Head on the south shore. This is most likely due to the tidal nature at Black Head as porpoises are known to use tidal races when feeding (Pierpoint, 2008). However, the visual dataset is limited as it is unable to provide a means to explore diel influence and has insufficient replication to explore tidal influences. SAM is necessary to fill these gaps and shows that porpoises are present at the site on an almost daily basis, 88% of days (total 333 days) using T-PODs and on 95% of days (total 572 days) using C- PODs. This information was gathered independent of weather conditions and darkness but is limited due to lack of information on abundance. However, from previous work carried out in

#%$ the bay over the mid-summer to autumn months, it is clear that Galway Bay supports an important population of harbour porpoise (Berrow et al., 2008), with an adult to calf ratio similar to other sites around Ireland. Through SAM further valuable information can be gathered on how animals use a monitored site by looking for behaviour characteristics in the dataset. Behavioural analyses from Spiddal show that porpoises are found to occur in all months but spring was found to have the lowest number of detections. Additionally exploring the train characteristics of porpoises at the site show it to be regularly used as a feeding location (41% (60,386 trains) of the total click trains fell under the category foraging), especially during the winter months during night-time hours. It is important to note that during the C-POD study 2009-2010, the wave energy device was in place at the OE Test Site and hence facilitated an experiment under the PReCAST programmed to determine if this artificial structure had an impact on presence. Results failed to show a significant difference in detections between sites suggesting the OE platform did not influence harbour porpoise presence. Clearly the area at Spiddal is an important habitat for the Harbour porpoise with the almost daily presence at the site. This presence is influenced by seasonal, diel and tidal factors. As harbour porpoises (Annex II species of the Habitats Directive) are present throughout the year and entitled to strict habitat protection, care must be taken to ensure this development does not degrade this habitat or cause undue disturbance. These visual SAM results will serve to inform protocols of best practice for the area if work is to go ahead and thus ensure the presence of small cetaceans in the area is not negatively impacted upon. Mitigation measures

should take into account the acoustic disturbance of marine mammals at the site as cable ' laying will occur both within the water and on land, and associated noise input should be reviewed to minimise displacement and to prevent habitat exclusion or hearing impacts such TTS or PTS. Mitigation measures for the site should include: 1. Presence of a trained experienced Marine Mammal Observer to implement the NPWS best practice guidelines when all work is taking place and to implement appropriate buffer zones in good sea-state. 2. Target work to take place when porpoise presence is at its lowest, e.g. during the spring or early summer. 3. Only carry out observations during daylight hours. 4. Carryout SAM at the site during and after installation works to assess if avoidance behaviour is recorded and if so for how long it persists.

With appropriate mitigation it is unlikely that work at the site over a short duration will have an impact on harbour porpoises through habitat exclusion or noise impacts.

& & 5 REFERENCES

Allen, M.C.; Read, A.J.; Gaudet, J. and Sayigh, L.S. (2001) Fine-scale habitat selection of foraging bottlenose dolphins Tursiops truncatus near Clearwater, 222, 253–264.

Andersen, R.J. 1904 The teeth of Mesoplodon hectori. Irish Naturalists' Journal, 13, 126-127.

Berrow, S. D. and Rogan, E. 1997 Cetaceans stranded on the Irish coast, 1901-1995. Mammal Review, 27(1), 51-76.

Berrow, S. D., Whooley, P. and Ferriss, S. 2002 Irish Whale and Dolphin Group Cetacean Sighting Review (1991-2001). Irish Whale and Dolphin Group, 34pp.

Berrow, S.D., O’Brien, J. and Massett, N. 2006 Humpback whale Megaptera novaengliae. Irish Naturalists’Journal, 28(2), 339-340.

Berrow, S.D., Hickey, R., O’Brien, J. O’Connor, I. and McGrath, D. 2008 Harbour Porpoise Survey 2008. Report to the National Parks and Wildlife Service. Irish Whale and Dolphin Group, 31pp.

Berrow, S.D., O’Brien, J., O’Connor, I. and McGrath, D. 2009 Abundance estimate and acoustic monitoring of harbour porpoise Phocoena phocoena in the Blasket Islands candidate Special Area of Conservation. Biology and Environment, Proceeding of the Royal Irish Academy, 109B.

Bolt, H.E.; Harvey, P.V.; Mandleberg, L. and Foote, A.D. (2009) Occurrence of killer whales in Scottish inshore waters: temporal and spatial patterns relative to the distribution of declining harbour seal populations. Aquatic Conservation, 19, 671–675.

Cabot, D. 1967 A Sperm Whale Physeter catodon (L.) on Achill Island, Co. Mayo, March, 1967. Irish Naturalists Journal, 15(11), 326pp.

Carlström, J (2005) Diel variation in echolocation behaviour of wild harbour porpoises. Marine Mammal Science 21, 1, 1-12.

(*) Canning, S.J.; Santos, M.B.; Reid, R.J.; Evans, P.G.H.; Sabin, R.C.; Bailey, N. and Pierce, G.J. (2008): Seasonal distribution of white-beaked dolphins (Lagenorhynchus albirostris) in UK waters with new information on diet and habitat use. Journal of the Marine Biological Association of the United Kingdom 88, 6, 1159-1166.

Cooke, D. 1990 Bottle-nosed dolphin Tursiops truncatus (Montagu). Irish Naturalists Journal, 23(8), 332.

Cox, T.M.; Read, A.J.; Solow, A. and Tregenza, N.J.C. (2001) Will harbour porpoises (Phocoena phocoena) habituate to pingers? Journal of Cetacean Research and Management, 3(1), 81-86.

Clarke, R. 1982 An index of sighting conditions for whales and dolphins. Report to the International Whaling Commission, 32, 559-561.

Dähne, M.; U.K. Verfuß, Diederics; A.; Meding, A. and Benke, H. (2006) T-POD test tank calibration and field calibration. Static Acoustic Monitoring of Cetaceans, European Cetacean Society, Gydnia, 2006. ECS newsletter no.43-Sepcial Issue, July 2006, 1-55.

De Bhaldraithe, P. 1977 The distribution of Euphausiid larvae in Galway Bay, west of Ireland, with special reference to Nyctiphanes couchi (Bell). Proceeding of the Royal Irish Academy, 77(8).

D’Arcy Thompson, W. 1900 Risso’s dolphin, a cetacean new to the Irish fauna. The Irish Naturalist, VIII, 88.

Evans, P.G.H. and Hammond, P.S. 2004 Monitoring Cetaceans in European Waters. Mammal Review, 34 (1), 131-156.

Fairley, J.S. 1979 Pilot whale ashore at Barna Co. Galway. Irish Naturalists Journal, 19(11), 405.

Fairley, J.S and Dawson, B. 1981 White-sided dolphin stranded at Galway. Irish Naturalists Journal, 20, 350.

Fairley, J.S. and Mooney, E.P. 1985 Pygmy Sperm Whale Kogia Breveceps (BLAINVILLE). Irish Naturalists Journal, 22(4), 164.

+ ,

Fairley, J.S. 1998 Minke whale Balaenoptera acutostrata Lacepede. Irish Naturalists Journal, 26(3-4), 123.

Fairley, J.S. and MacLoughlin, E. 1990 Euphrosyne Dolphin Stenella coeruleoalba (Meyen). Irish Naturalists Journal, 23(8), pp 336.

Fernandes, L.J.M. 1988 A study of the oceanography of Galway Bay, mid-western coastal waters (Galway Bay to Tralee Bay), Shannon Estuary and River Shannon plume. Department of Oceanography, University College Galway. Ph.D Thesis, 229pp.

Fraser, F.C. (1934, 1946, 1953, 1974) Reports on cetacean stranded on the British coasts from 1927 to 1966, 11-14. British Museum (Natural History), London.

Gilles, A.; Adler, S.; Kaschner, K.; Scheidat, M. and Siebert, U. (2011) Modelling harbour porpoise seasonal density as a function of the German Bight environment: implications for management. Endangered Species Research 14, 157-169.

Harmer, S.F. (1914-27) Reports on Cetacea stranded on British coasts. British Museum (Nat. Hist.) London, 1-10.

Kyhn, L.A. (2006) Detection of harbour porpoises using T-PODs. M.Sc Thesis, National Environmental Research Institute, Denmark, 1-58.

Kyhn, L.A.; Tougaard, J.; Teilmann, J.; Wahlberg, M.; Jørgensen, P.B. and Bech, N.I. (2008) Harbour porpoise (Phocoena phocoena) static acoustic monitoring: laboratory detection thresholds of T-PODs are reflected in field sensitivity. Journal of the Marine Biological Association of the United Kingdom, 88, 1085-1091.

Lei, W. 1995 3-Dimensional hydrographic modeling in Galway Bay. Ph.D Thesis. Department of Oceanography, University College Galway.

Mate, B.R.; Rossbach, K.A.; Nieukirk, S.L.; Wells, R.S.; Irvine, B.A.; Scott, M.D. and Read, A.J. (1995) Satellite monitored movements and dive behaviour of a bottlenose dolphin in Tampa Bay, Florida. Marine Mammal Science 11, 4, 452-463.

-*. Marubini, F.; Gimona, A.; Evans, P.G.H.; Wright, P.J. and Pierce, G.J. (2009) Habitat preferences and interannual variability in occurrence of the harbour porpoise (Phocoena phocoena) off northwest Scotland. Marine Ecology Progress Series, 381, 5, 297–310.

McGrath, D. 1983 Killer whale Orchinus orca (L). Irish Naturalists Journal, 21, 332.

Moffat, C.B. 1938 The mammals of Ireland. Proceedings of the Royal Irish Academy, 44B, 61– 128.

Nelson, P.A. (2003) Marine fish assemblages associated with fish aggregating devices (FADs): effects of fish removal, FAD size, fouling communities, and prior recruits. Fish. Bull. 101: 835- 850.

Nolan, G.D. 1997 A study of the River Corrib plume and its associated synamics in Galway Bay during the winter months. Department of Oceanography, National University Of Ireland, Galway. Ph.D Thesis, 94pp.

O’Brien, J. (2009) The inshore distribution and abundance of small cetaceans on the west coast of Ireland: Site assessment for SAC designation and an evaluation of monitoring techniques. Galway-Mayo Institute of Technology, Unpublished Ph.D Thesis, pp 1-226.

O’Brien, J., McGrath, D. and Berrow, S. 2006 Bottlenose dolphins and harbour porpoises in Galway Bay and North Connemara. Proceedings from the 20th Annual European Cetacean Society Conference, Poland, 2006.

O’Brien, J., Beck, S., Berrow, S.D. and Hansen, S. (Submitted) Detection ability and calibration of C-POD and T-POD units off the West coast of Ireland. Journal of Marine Biological Association.

O’Brien, J., Beck, S., Wall, D., Pierini, A., and Hansen, S. (2012). Marine Mammals and Megafauna in Irish Waters, behaviour, distribution and habitat use. Work Package 2: Developing Acoustic Monitoring Techniques. PReCAST Final Report – Marine Research Sub- programme 2007-2013, pp1-205.

O'Cadhla, O., Mackey, M., Aguilar de Soto, N., Rogan, E., and Connolly, N. 2004 Cetaceans and Seabirds of Ireland's Atlantic Margin. Volume II-Cetacean distribution & abundance.

/ 0 Report on research carried out under the Irish Infrastructure Programme (PIP): Rockall Studies Group (RSG) projects98/6 and 00/13, Porcupine Studies Group project P00/15 and Offshore Support Group (OSG) project 99/38, 1- 82.

O'Riordan, C.E. 1972 Provisional list of cetacea and turtles stranded or captured on the Irish coast. Proceedings of the Royal Irish Academy, 72(15), 253-274.

O'Riordan, C.E. 1976 Marine fauna notes from the National Museum of Ireland. Irish Naturalists Journal, 18, 349-350.

Palka, D. 1996 Effects of Beaufort sea state on the sightability of harbour porpoises in the Gulf of Maine. Rep. Int. Whal. Commn, 46, 575-582.

Philpott, E.; Englund, A.; Ingram, S. and Rogan, E. (2007) Using T-PODs to investigate the echolocation of coastal bottlenose dolphins. Journal of Marine Biological Association, 87, 11-17.

Read A.J. and Westgate A.J. 1997 Monitoring the movements of harbour porpoises (Phocoena phocoena) with satellite telemetry. Mar Biol, 130,315–322.

Pierpoint, C. 2008. Harbour porpoise (Phocoena phocoena) foraging strategy at a high energy, near-shore site in south-west Wales, UK. Journal of Marine Biological Association, 88(06), 1167-1173.

Scharff, R.F. 1900 A list of Irish Cetacea. Irish Naturalist, 9, 83-91.

Shane, S.H. 1990 Behaviour and ecology of the bottlenose dolphin at Sanibel Island, Florida. In S.Leatherwood & Randall R. Reeves (Eds.). The bottlenose dolphin. San Diego: Academic Press, 245-265.

Smolker, R.A., Richards, A.F., Connor, R.C. and Pepper, J.W. 1993 Sex differences in patterns of association among Indian Ocean Bottlenose Dolphins. Behaviour 123, 38-69.

Simon, M., Nuuttila, H., Reyes-Zamudio, M.M., Ugarte, F., Verfuß, U.K. and Evans, P.H. (2010) Passive acoustic monitoring of bottlenose dolphin and harbour porpoise with implications for habitat use and partitioning. Journal of the Marine Biological Association of the United Kingdom 90, 8, 1-7. 13

1*2

Teilmann, J. 2003 Influence of sea state on density estimates of harbour porpoises (Phocoena phocoena). J. Cetacean Res. Manage, 5(1):85-92.

Todd, V.L.G., Pearse, W.D., Tregenza, N., Lepper, P.A. and Todd, I.B. (2009) Diel echolocation activity of harbour porpoises (Phocoena phocoena) around the North Sea offshore gas installations. ICES Journal of Marine Science 66, 4, 734-745.

Tougaard, J.; Poulsen, LR; Amundin, M; Larsen, F; Rye, J. and Teilman J. (2006) Detection function of T-PODs and estimation of porpoise densities. Static Acoustic Monitoring of Cetaceans. Proceedings of the European Cetacean Society, Gdnyia, Poland. Special Issue, 46: 7- 14. 20th Annual Meeting, April 2006.

Villadsgaard, A; Wahlberg, M and Tougaard, T (2007) Echolocation signals of wild harbour porpoises, Phocoena phocoena. Journal of Experimental Biology 210, 1, 56-64.

Young, R.F., and Peace, S. 1999 Using simultaneous counts by independent observers to correct for observer variability and missed sightings in a shore-based survey of bottlenose dolphins. J. Cetacean Res. Manage, 1(3), 279-287.