Supplementary Material A: Methods & results for sound ...
Supplementary Material for ‘MULTIPLE FIELD-BASED METHODS TO ASSESS THE POTENTIAL IMPACTS OF SEISMIC SURVEYS ON SCALLOPS’Rachel Przeslawski, Zhi Huang, Andrew Carroll, Matthew Edmunds, Peter D. Nichols, Stefan Williams REF _Ref482871779 \h \* MERGEFORMAT Supplementary Material A: Methods & results for sound monitoring REF _Ref482871802 \h \* MERGEFORMAT Supplementary Material B: Methods & results for sound propagation modelling REF _Ref482871806 \h \* MERGEFORMAT Supplementary Material C: Scallop dredging method photos REF _Ref482871811 \h \* MERGEFORMAT Supplementary Material D: Details of AUV methodsSupplementary Material E: Results of conductivity-temperature-depth casts REF _Ref482871816 \h \* MERGEFORMAT Supplementary Material F: Results from dredging and AUV dataSupplementary Material A: Methods & results for sound monitoring Hydrophone Calibration DetailsHydrophones have undergone two stages of calibration, the first stage by the manufacturer and the second stage by the ARU provider. For the first stage, the hydrophone manufacturer (Wildlife Acoustics) calibrated the hydrophones in batch prior to their installation in the Acoustic Recording Units (ARUs) at the National Physical Laboratory in Plymouth, UK. A chart representing the calibration curves from the batch of hydrophones including those installed into the ARUs for the current study is shown in REF _Ref482881480 \h \* MERGEFORMAT Figure 1, and the calibrations curves of the associated reference hydrophone is shown in REF _Ref482881510 \h \* MERGEFORMAT Figure 2.Figure SEQ Figure \* ARABIC 1 Calibration curves provided by manufacturer for hydrophone model HTI-99-HF at 2.3 m depth and 19.2°CFigure SEQ Figure \* ARABIC 2Calibration curve of reference hydrophone used to calibrate model HTI-99-HF hydrophone.For the second stage, the ARU provider (Gardline) tested the ARU by generating a known steady state reference test tone at 1 kHz, recording the test tone using the ARU for 30 seconds, and converting the measured voltage to determine the sensitivity ( REF _Ref482881570 \h \* MERGEFORMAT Figure 3). It was assumed the hydrophone exhibits linear behaviour at the frequency outside its resonance frequency, and has +/- 3 dB operation accuracy at its operating frequency. A calibration data point was provided by the manufacturer and was linearly interpolated based on the manufacturer’s reference calibration ( REF _Ref482881510 \h \* MERGEFORMAT Figure 2) (Hayman et al. 2016). Figure SEQ Figure \* ARABIC 3 Calibration curves for ARUs used in the current study, showing a sensitivity test against a reference hydrophone.In the case of low frequency below 1kHz, the frequency response were extrapolated using data provided by the specification, which has a roll off of approximately -15dB at 25- 100Hz. With each ARU unit, a correction curve was generated based on the measured sensitivity and was applied accordingly during data processing for the error correction using a pistonphone (Hayman et al. 2016, RESON Group, ). Sound Monitoring MethodsFour acoustic recording units (ARUs) were moored to the seabed before the seismic survey and collected afterwards, thus measuring benthic sound exposure before, during and after airgun operations. The ARUs used were shallow and deep water Song Meter SM2M+ Marine Recorders, submersible 16-bit digital recorders with a dynamic range of 78-165 dB (sound pressure level), and an audio sampling rate of 4-96 kHz. Upon deployment, each ARU was completely submerged and positioned a minimum of 35 m below the surface to avoid entanglement with seismic equipment and using acoustic release transponders to allow moorings to be recovered ( REF _Ref433372081 \h \* MERGEFORMAT Figure 4). The transponders were activated by an acoustic signal from the surface vessel, causing them to release the mooring and float to the surface. To enable recovery a rope was attached to the ballast, which was sufficiently long to bridge the gap between the subsurface buoy and the seabed. All exact locations of the deployments were logged using a handheld GPS. ARUs were deployed at three sites in the experimental zone, as well as a control site over 25 km from seismic airgun shots ( REF _Ref433798490 \h \* MERGEFORMAT Figure 5). Locations were chosen to be adjacent to planned AUV missions and to represent a range of depths.Figure SEQ Figure \* ARABIC 4: Schematic of ARU deployment, not shown to scale.Figure SEQ Figure \* ARABIC 5: Map of sound monitoring and modelling sites overlaid on seismic survey lines. Number indicate seismic lines in which sound monitoring was analysed (See Results - REF _Ref433798440 \h \* MERGEFORMAT Sound Monitoring below). ARUs were deployed at all four sites and associated data modelled. Theoretical sound modelling was conducted only at the three experimental sites (A, B, and C).The ARUs were deployed on 3-4 April. During seismic operations, they were initially set to record for 2 minutes with a 10 second pause. However, this proved insufficient for data storage, and a service was necessary on 14-15 April to adjust each recording to 1 minute duration with a 10 second pause. Sound from airgun operations were recorded by the hydrophones at all four sites between 14-15 and 20 April. There was a malfunction of one of the hydrophones (ARU B) between 3 and 14 April, and data was therefore only collected from three hydrophones during this time. Although ARU B failed to record sound when the airguns were operating directly overhead, it did record sound when the airguns were operating nearby (< 2 km). As such, we have reliably measured and modelled sound exposure at this site (as well as 3 others) which fully meets the proposal objectives. ARU data was analysed by calculating the measured level in terms of the sound exposure level (SEL), expressed in one-third octave band (TOB) units, relative to 1 μPa from 20 Hz to 24 kHz. The Sound Exposure Level (SEL) is a measure of the pulse energy content. The SEL for a single pulse is calculated by integrating the square of the pressure waveform over the duration of the pulse. The duration of the pulse is defined as the region of the waveform containing the central 90% of the energy of the pulse. The calculation is given by the following equation where p = pressure and t = time: The value is then expressed in dB re 1 ?Pa2s and is calculated from the equation below where E0 is the reference value of 1 ?Pa2s:Sound Monitoring ResultsARUs prior to the commencement of the seismic survey. Background noise data for Gippsland Basin is presented as power spectral density levels (dB re 1 ?Pa2/Hz) in one-third octave bands (TOB) expressed from 10 Hz to 24 kHz ( REF _Ref444869062 \h \* MERGEFORMAT Figure 6). The primary contributors to background sound levels in the Gippsland Basin were wind, rain, and current- and wave-associated sound. Biological sounds, including dolphin vocalisations, were also recorded. There was considerable variation in background sound for all four locations, although low frequencies (<2 kHz) dominated. Sites C and D are subject to lower background sound levels (at 100 – 500 Hz) than the sites A and B ( REF _Ref433878708 \h \* MERGEFORMAT Table 1), likely due to increased distance from shipping activity (Site D) and deeper water associated with fewer acoustic echoes (Site C).Figure SEQ Figure \* ARABIC 6: Spectral density levels of background sound measured in one third octave band (TOB) at each of the four hydrophone sites. Different colour lines represent different dates and times from 14 – 20 April 2015.Table SEQ Table \* ARABIC 1: Maximum background spectral density levels recorded at each ARU SiteTreatmentDepthMax background sound (dB re 1 ?Pa2/Hz)aAExperimental41109.9BExperimental51104.6CExperimental68100DControl5289.2Analysis of sound associated with airgun operations revealed that the primary frequency was between 20 and 500 Hz. Two seismic lines were chosen to monitor sound exposure levels at the ARUs (Lines 1024 and 1032). The survey lines were selected at reasonable distance away from all measuring ARUs to ensure far field assumptions for acoustic propagation modelling, in order to reduce inconsistency due to near field acoustic interference at close range. Line 1024 and 1032 also gave the most comprehensive data set that was captured by all 4 recording stations without clippings and malfunctioning. A further two lines were chosen at which only the closest point of the vessel was analysed (Lines 1027 and 1033) in order to provide the shortest distance possible between the sound source and receiver (see REF _Ref433798490 \h \* MERGEFORMAT Figure 5 for map of line locations). The maximum sound exposure levels (SELs) for these lines, as recorded by the ARUs, are shown in REF _Ref433805897 \h \* MERGEFORMAT Table 2. For each line, these represent the highest sound levels for each line where the airgun array was producing impulsive sound at its maximum output and the ARU was recording. Among these lines, the highest SEL recorded was 146 dB re 1 ?Pa2s at Site B (51 m depth) when the airguns were operating 1.4 km away. Maximum received sound levels were lower at greater depths even when horizontal distance was less due to sound attenuation in water (e.g. Site A vs Site D along Line 1032 in REF _Ref433805897 \h \* MERGEFORMAT Table 2). Airgun sound signals were detected up to 60 km away (e.g. Site A REF _Ref433805897 \h \* MERGEFORMAT Table 2). Sample spectrograms and amplitudes for Lines 1024 are shown in REF _Ref433798798 \h \* MERGEFORMAT Figure 7 and REF _Ref433878831 \h \* MERGEFORMAT Figure 8.Table SEQ Table \* ARABIC 2: Received sound levels recorded from the ARUs during seismic survey lines 1024, 1032, 1027, and 1033. Only the nearest ARU was analysed for Lines 1027 and 1033. See REF _Ref433798490 \h \* MERGEFORMAT Figure 5 for site and line locations.SiteDepth (m)Sediment grain-size (% mud, sand, gravel)aSeismic LineHorizontal distance to acoustic source (km)SEL (dB re 1 ?Pa2s)bA414, 84, 13102438.7107.1B515, 77, 18102420.1124.3C685, 82,1410248.2135.4D5220, 65, 14102451.393.7A414, 84, 13103260.0123.6B515, 77, 18103237.3121.4C685, 82,14103219.5120.9D5220, 65, 14103238.1113.7C685, 82,1410270.2140.3 c,dB515, 77, 1810331.4146.0ea Sediments collected from shipek grab. At Sites A and C, one grab each was deployed, while at Sites B and D four grabs each were deployed (mean grain-size presented for Sites B and D). b Unweighted SEL TOBc Estimated airgun pulse duration: 0.4068 secd These values are lower than those known for a similar size airgun array at that distance (~170 dB re 1 ?Pa2s) (Day et al. 2016 and references therein), and it is likely that the signals overloaded the recorders at the short range of 200 m. This value is therefore not considered accurate.e Estimated airgun pulse duration: 0.3925 secFigure SEQ Figure \* ARABIC 7: Sample spectrograms for airgun pulses along seismic line 1024 at all four sites at which ARUs were deployed. See REF _Ref433798490 \h \* MERGEFORMAT Figure 5 for site locations.Figure SEQ Figure \* ARABIC 8: Sample amplitudes for airgun pulses along seismic line 1024 at all four sites at which ARUs were deployed. See REF _Ref433798490 \h \* MERGEFORMAT Figure 5 for site locations.Supplementary Material B: Methods & results for sound propagation modellingSound Modelling MethodsSource Level is a metric used in underwater acoustics to describe the source output amplitude. The decibel units for this quantity may be written as dB re 1 μPa2m2. It should be noted that source level is an idealised acoustic far-field parameter and is not necessarily equal to the acoustic pressure or received level measured at a distance of 1 m from the source. It may be considered as the sound pressure level that would exist at a range of 1 m from the acoustic centre of an equivalent simple source, which radiates the same acoustic power into the medium as the source in question. In addition, seismic arrays are directional systems that are designed to direct the majority of the energy towards the seabed. The directionality of the noise source has not been included in the model as the modelling locations are static, while in reality the survey is dynamic. It is estimated that directionality could result in received noise source levels fluctuating by up to 10 dB depending on the relative bearing of the receiver to the seismic array.Table SEQ Table \* ARABIC 3: Sound propagation modelling characteristicsContractorSVT Engineering ConsultantsModelled responseReceived level (SEL unweighted)Number of sites3 (Sites A, B, C)Source 16-gun 2530 cubic inch array, as provided by seismic vessel operator Source Level215 dB re 1 μPa2s aSource Height6 m below sea surfaceReceived LevelTo be estimated by modelReceiver Height 1 m above seabedModel categoryParabolic EquationModel typeMonterey Miami Parabolic Equation (MMPE)Model limitationsRough surface scatteringb, vertical launch anglec, omnidirectionalitydEnvironmental inputs Tides/wave height3.1 m (tide)e Substrate typeMedium sandg Sound speed profileFrom CTD data ( REF _Ref433813664 \h \* MERGEFORMAT Table 4)a As derived from the time domain signal provided by seismic vessel operatorb The model is conservative because acoustic wave scattering due to the roughness of the sea surface/seabed is not accounted forc The launch angle of the model is limited to ±40°. The sound waves predicted at angles close to the noise source outside of this angle are evanescent waves, i.e. strongly decayingd The model conservatively assumes an omni-directional source that does not take directionality into account. Seismic airguns are directional sources, and the measured level is therefore dependent on the relative angle of the seismic survey to the receiver.e Highest recorded level recorded at Port Welshpool (Bureau of Meteorology) for worst case scenariof Tide information from Global Ocean Tide Model ()g Based on GA’s Marine Sediments (MarS) database and assuming unconsolidated sub-surface geology. Sound speed of 1774 m/s, density of 2.05 g/cm3, sound attenuation of 0.374 dB/m/kHz from Richardson and Briggs 2004.h Based on grain size from grab samples collected at sites in June 2015 and assuming unconsolidated sub-surface geology. Sound speed of 1893 m/s, density of 2285 kg/cm3, sound attenuation of 0.87 dB/λ from Ainslie 2010.Modelling studies require temperature, salinity, and depth data to account for the dependence of sound propagation through water on these factors. To this end CTD (conductivity-temperature-depth) meters were deployed at each of the four sites ( REF _Ref433798490 \h \* MERGEFORMAT Figure 5). Data from the CTD ( REF _Ref433813664 \h \* MERGEFORMAT Table 4) was then used for sound speed profiling during propagation loss modelling.Table SEQ Table \* ARABIC 4: Data obtained from conductivity-temperature-depth (CTD) metersSite ASite BSite CSite DMean sound speed (m/s)1515.21514.11516.61514.4Mean temperature (°C)17.3817.0817.8417.2Max water depth of CTD cast (m)44.154.867.955.4Particle velocity levels received near the seabed were estimated because invertebrates are more likely to detect particle motion rather than pressure ADDIN EN.CITE <EndNote><Cite><Author>Hawkins</Author><Year>2015</Year><RecNum>579</RecNum><DisplayText>(Hawkins et al., 2015)</DisplayText><record><rec-number>579</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">579</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hawkins, A. D.</author><author>Pembroke, A.E.</author><author>Popper, A. N.</author></authors></contributors><titles><title>Information gaps in understanding the effects of noise on fishes and invertebrates</title><secondary-title>Reviews in Fish Biology and Fisheries</secondary-title></titles><periodical><full-title>Reviews in Fish Biology and Fisheries</full-title></periodical><pages>39-64</pages><volume>25</volume><dates><year>2015</year></dates><urls></urls></record></Cite></EndNote>(Hawkins et al., 2015). The acoustic pressure gradient was estimated using the field data acquired from the ARUs and the Finite Difference Method ADDIN EN.CITE <EndNote><Cite><Author>LeVeque</Author><Year>2007</Year><RecNum>657</RecNum><DisplayText>(LeVeque, 2007)</DisplayText><record><rec-number>657</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">657</key></foreign-keys><ref-type name="Book">6</ref-type><contributors><authors><author>LeVeque, R.</author></authors></contributors><titles><title>Finite Difference Methods for Ordinary and Partial Differential Equations: Steady- state and Time Dependent Problems</title></titles><pages>328</pages><dates><year>2007</year></dates><publisher>Society for Industrial and Applied Mathematics</publisher><urls></urls></record></Cite></EndNote>(LeVeque, 2007). Particle velocity was analysed at four frequencies (40, 80, 160, 200 Hz), and particle velocity fields (i.e. throughout the entire water column and shallow seabed) were calculated based on the modelled transmission loss from RAMSGeo. The particle velocity field was then used to estimate single strike particle velocity levels (i.e. received level at a single point near seafloor) at increasing distances from the acoustic source.Sound Modelling ResultsThe maximum SEL received one metre above the seafloor with airguns directly overhead was predicted to be 170 dB re 1 μPa2s, extending 200 – 250 m from the receiver depending on water depth and directionality. Predicted SELs greater than 150 dB re 1 μPa2s extended almost 4 km from the receiver, with predicted SELs of 125-130 dB re 1 μPa2s extending out to 30 km.Particle velocity was predicted to be higher at lower frequencies than the higher frequencies for Lines 1024 and 1032 ( REF _Ref438038555 \h \* MERGEFORMAT Figure 9). For Line 1024 the highest modelled peak particle velocity is 171 dB re 1 nm/s at 40 Hz, and for Line 1032 the modelled peak particle velocity is 166.7 dB re 1 nm/s at 80 Hz ( REF _Ref433894115 \h \* MERGEFORMAT Table 5).Figure SEQ Figure \* ARABIC 9: Modelled particle velocity field (mm/s) throughout the water column at various frequencies and distances from sound source for a) seismic line 1024 and b) seismic line 1032. Black lines indicate the seabed.Table SEQ Table \* ARABIC 5: Modelled peak particle velocity (dB re 1 nm/s) at various frequencies with increasing distance from the sound source.Distance from acoustic source (m)Line 10241003005001000 40 Hz171.0161.6155.7152.3 80 Hz162.8151.0152.3147.9 160 Hz123.4119.3116.4112.4 200 Hz128.8125.9119.1116.1Line 1032 40 Hz157.7150.8142.4137.4 80 Hz166.7160.5154.6145.5 160 Hz136.8131.4131.3121.7 200 Hz135.7138.6129.8128Supplementary Material C: Scallop dredging method photosFigure SEQ Figure \* ARABIC 10 Examples of different gonad stages as per classification system described in Young et al. (1999) and Harrington et al. (2010). Arrows point to scallop gonad.Figure SEQ Figure \* ARABIC 11 Metrics recorded from photographs taken of scallops collected from dredges, including both unopened (a) and opened (b) scallops. Doughboy scallops were not opened, and therefore only height and length data are available.Supplementary Material D: Details of AUV methodsThe AUV Phoenix is jointly operated by Australian Maritime Ecology Pty Ltd and the University of Sydney’s Australian Centre for Field Robotics (ACFR). The AUV is an OceanServer Iver-2 model (~35 kg) with an in-built stereo camera, strobe and modem systems. The navigation system includes a GPS (when on the surface), compass, four-beam Doppler velocity log and ultra-short baseline (USBL) positioning system. Images were collected with high-sensitivity CCD cameras (Allied Vision Manta G-145C) with a resolution of 1388 x 1038 pixels. High-resolution, georeferenced stereoscopic images of the seafloor were collected at a rate of 2 Hz with the AUV travelling at a target altitude of 2 m above the seabed at a speed of 1 m s-1 (2 knots). Illumination was by two strobes mounted in the fore and aft-sections of the vehicle and synchronised with the cameras. The camera lens and stereo orientation properties were calibrated in a pool immediately prior to each survey.The AUV Sirius (~200 kg) is operated by the ACFR and is a modified version of a robotic vehicle called Seabed (whoi.edu). A variety of navigational sensors including GPS, Ultra Short Baseline Acoustic Positioning System (USBL) and forward looking obstacle avoidance sonar, enable precise tracking of the vehicle and allows survey data to be geo-referenced at high precision. Images were collected with a synchronized Images were collected with a synchronized pair of high sensitivity 12 bit, 1.4 megapixel cameras (AVT Prosilica GC1380 and GC1380C; one monochrome and one colour). Illumination was achieved by two 4-J strobes mounted in the fore and aft-sections of the vehicle and synchronised with the cameras ADDIN EN.CITE <EndNote><Cite><Author>Williams</Author><Year>2012</Year><RecNum>815</RecNum><DisplayText>(Williams et al., 2012)</DisplayText><record><rec-number>815</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">815</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Williams, Stefan B</author><author>Pizarro, Oscar R</author><author>Jakuba, Michael V</author><author>Johnson, Craig R</author><author>Barrett, Neville S</author><author>Babcock, Russell C</author><author>Kendrick, Gary A</author><author>Steinberg, Peter D</author><author>Heyward, Andrew J</author><author>Doherty, Peter J</author></authors></contributors><titles><title>Monitoring of benthic reference sites: using an autonomous underwater vehicle</title><secondary-title>IEEE Robotics & Automation Magazine</secondary-title></titles><periodical><full-title>IEEE Robotics & Automation Magazine</full-title></periodical><pages>73-84</pages><volume>19</volume><number>1</number><dates><year>2012</year></dates><isbn>1070-9932</isbn><urls></urls></record></Cite></EndNote>(Williams et al., 2012). The AUV mapped the seabed at an altitude of 2 m and a cruising speed of 0.5?m s–1.The AUV Phoenix was programmed to survey a 2750 m long two-loop pattern at each site ( REF _Ref483409501 \h \* MERGEFORMAT Figure 12). The survey pattern was designed with two cross-over points (loop closures) to assist in refining vehicle position estimates during data post-processing. The AUV deployment was from a single surface start and return point. During the February 2016 survey, one of the AUV Phoenix cameras failed, limiting the image acquisition to 2D imagery and restricting the mosaicking of the imagery. All individual images could still be georeferenced. The AUV Sirius was used at Sites 40 and 41 when images were unsuccessfully acquired from the AUV Phoenix. 1371603189605Figure SEQ Figure \* ARABIC 12 Standardised benthic survey pattern (‘snake-return’) used for the Gippsland Basin survey, June 2015 and February 2016. The direction of the main axis varied between sites.0Figure 12 Standardised benthic survey pattern (‘snake-return’) used for the Gippsland Basin survey, June 2015 and February 2016. The direction of the main axis varied between sites.134620271780300 mStartEndDeployment /Recovery00300 mStartEndDeployment /RecoveryData was extracted from images using the online annotation platform Squidle () which allowed two analysts (one analyst per survey) to categorise georeferenced images based on in situ observations of the seafloor. Approximately every 3-5 images were annotated, depending on the speed of the AUV, such that a continuous but non-overlapping series of images were annotated. Image annotation used the CATAMI classification scheme ADDIN EN.CITE <EndNote><Cite><Author>Althaus</Author><Year>2015</Year><RecNum>660</RecNum><DisplayText>(Althaus et al., 2015)</DisplayText><record><rec-number>660</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">660</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Althaus, Franziska</author><author>Hill, Nicole</author><author>Ferrari, Renata</author><author>Edwards, Luke</author><author>Przeslawski, Rachel</author><author>Sch?nberg, Christine H. L.</author><author>Stuart-Smith, Rick</author><author>Barrett, Neville</author><author>Edgar, Graham</author><author>Colquhoun, Jamie</author><author>Tran, Maggie</author><author>Jordan, Alan</author><author>Rees, Tony</author><author>Gowlett-Holmes, Karen</author></authors></contributors><titles><title>A Standardised Vocabulary for Identifying Benthic Biota and Substrata from Underwater Imagery: The CATAMI Classification Scheme</title><secondary-title>PLoS One</secondary-title></titles><periodical><full-title>PLoS One</full-title></periodical><pages>e0141039</pages><volume>10</volume><number>10</number><dates><year>2015</year></dates><publisher>Public Library of Science</publisher><urls><related-urls><url>;(Althaus et al., 2015) to identify substrate type (e.g. unconsolidated: sand/mud), bedforms (e.g. two-dimensional ripples), and animals of interest including scallops, other bivalves (e.g. screwshells, Limatula beds), and fish. Both commercial and doughboy scallops were counted and assigned modifiers based on the position of their valves (open, closed, indeterminate) and their location in the sediment (fully, partially, or un-buried) ( REF _Ref483409673 \h Figure 13)). Dead or disarticulated scallop shells were also scored, except for sites which had a high cover of shell debris. Categorisation of scallops based on these classifiers allows for the determination of scallop condition and overall health throughout each site: open and buried scallops were considered ‘live’ as this indicated a healthy recessed scallop with the upper valve level with the substrate ADDIN EN.CITE <EndNote><Cite><Author>Brand</Author><Year>2016</Year><RecNum>838</RecNum><DisplayText>(Brand, 2016)</DisplayText><record><rec-number>838</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">838</key></foreign-keys><ref-type name="Book Section">5</ref-type><contributors><authors><author>Brand, A.R.</author></authors><secondary-authors><author>Shumway, S.E.</author><author>Jay Parsons, G.</author></secondary-authors></contributors><titles><title>Scallop ecology: distribution and behaviour</title><secondary-title>Scallops: Bioogy, Ecology, Aquaculture, and Fisheries</secondary-title></titles><pages>469-513</pages><edition>3rd</edition><dates><year>2016</year></dates><pub-location>New York</pub-location><publisher>Elsevier</publisher><urls></urls></record></Cite></EndNote>(Brand, 2016). Disarticulated and dead shells were self-evident, and all other shell categories (partially buried/unburied, closed/indeterminate gape) were considered of unknown viability.Figure SEQ Figure \* ARABIC 13 Examples of digitally overlaid modifiers used to annotate scallops in images acquired from the AUV Phoenix (after the seismic survey). (Left) Commercial scallops (P. fumatus); (Right) Doughboy scallops (M. asperrima).For the short-term after survey (GA-353), a total of 6148 images were annotated for bedforms, substrate, scallops and other organisms. Numbers of images annotated varied among stations due to acquisition issues (sites 12, 36, 40, 41) or time available for annotation (Stations 06, 14) ( REF _Ref438115114 \h Table 6). However, numbers of annotated images were more balanced among treatments (2802 in control and 3346 in experimental zones), and nested multivariate analyses allowed an unbiased statistical assessment of differences within and among sites. For the long-term after survey (GA-355), a total of 2420 images were annotated for bedforms, substrate, scallops and other organisms. Total number of images annotated is less than the short term after survey (GA-353) due to time available for annotation. Raw images have been released online ( REF _Ref483409700 \h Table 7).Table SEQ Table \* ARABIC 6: Numbers of images annotated from the surveys GA-353 (short-term after) and GA355 (long-term after). Numbers in parentheses indicate total number of scallops (C = Commercial Scallops; D = Doughboy Scallops). Control refers to sites > 10 km from airgun operations, and experimental refers to sites 0-1 km from airgun operations. Depth derived from annotated AUV images (rounded to nearest m)SiteTreatmentDepth rangeNo. images analysed (short-term)No. images analysed (long-term)05Control371569 (641C; 373D)298 (44C; 4D)06Control4473 (6C; 4986D)48 (2C; 497D)08Experimental521284 (51C; 85D)190 (13C; 140D)12Control54846 (198C; 22D)133 (21C; 42D)14Control59574 (6C; 154D)122 (13C; 122D)36Control3749 (34C; 1D)268 (81C; 0D)37Control60469 (381C; 1501D)204 (202C; 192D)40Experimental61337 (21C; 2D)401 (63C; 50D)41Experimental60246 (1C; 4D)513 (29C; 246D)45Experimental531479 (212C; 7543D)246 (36C; 771D)Table SEQ Table \* ARABIC 7: Publicly released data collected from the AUV-Phoenix and AUV-Sirius and links to their online locations. ‘Short-term’ survey refers to data collected 20-24 June 2015, and Long-term’ refers to data collected 20-22 February 2016. Images and environmental data acquired from the ‘before’ survey are also available online (see table 10 in ADDIN EN.CITE <EndNote><Cite><Author>Przeslawski</Author><Year>2016</Year><RecNum>807</RecNum><DisplayText>(Przeslawski et al., 2016)</DisplayText><record><rec-number>807</rec-number><foreign-keys><key app="EN" db-id="arp55rdtqwd0adetva4p299cdxwxrvpx502w">807</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Przeslawski, R.</author><author>Hurt, L.</author><author>Forrest, A.</author><author>Carroll, A.</author></authors></contributors><titles><title>Potential short-term impacts of marine seismic surveys on scallops in the Gippsland Basin</title></titles><pages>60</pages><number>FRDC Report 2014/014</number><dates><year>2016</year></dates><pub-location>Canberra</pub-location><publisher>Fisheries Research and Development Corporation</publisher><urls></urls></record></Cite></EndNote>(Przeslawski et al., 2016).Data TypeDescriptionSurveyData linkStill Images.tiff files from AUV-Phoenix. GeoreferencedShort-term Images.tiff files from AUV-Phoenix and AUV-Sirius. GeoreferencedLong-term Mosaics.tiff files from AUV-SiriusLong-term MeshTiles for GIS importLong-term Material E: Results of conductivity-temperature-depth castsWe used conductivity-temperature-depth (CTD) data to determine whether vertical mixing was strong enough to allow applicability of our SST modelling to water near the seabed. CTD data was obtained from casts undertaken by Gardline Marine Science Pty Ltd to inform the sound modelling component of the project (see Appendix A). CTD data were analysed at four sites (A-D) in the Gippsland Basin during two occasions in April 2015 ( REF _Ref449619070 \h \* MERGEFORMAT Figure 14). Figure SEQ Figure \* ARABIC 14: Location of CTD casts in April 2015 in the Gippsland Basin. Map created in ArcMap 10.2.The results of the CTD casts show that water temperature was remarkably consistent throughout the water column at most sites ( REF _Ref475531178 \h Figure 15). The deeper water sites (C and D) showed the most variation in thermal vertical profiles on 14 April 2015, but this still only encompassed a range of < 0.15°C. Salinity was also consistent throughout the water column, although there was some evidence of salinity strata at the sea surface and seabed that showed temporal and spatial variation. These CTD results indicate that, in the Bass Strait area, the spatial patterns of water temperature observed on the surface can be inferred into similar patterns of water temperature near the seabed. Figure SEQ Figure \* ARABIC 15: Depth profiles of temperature (top panel) and salinity (bottom panel) at four sites in the Gippsland Basin (see Figure 1), each of which shows two time periods.Supplementary Material F: Results from dredging and AUV dataFigure SEQ Figure \* ARABIC 16 Shell assemblages from P. fumatus collected on short-term (pale colour) and long-term (darker colour) surveys. This is a sub-set of the dredges shown in Figure 3 in the main text and only include those tows from areas that were consistently sampled among times. See Figure 2 in main text for dredge locations.Table SEQ Table \* ARABIC 8: Results of nested PERMANOVA tests on scallop metrics derived from onboard photographs including a) height, b) length, c) AM diameter, d) ovary area, e) testis area, f) gonad (i.e. roe) area, and g) gonad stage.Seismic impacts would have been indicated by a significant interaction between zone and time (zone * time). Bold text represents significant relationships at α = 0.05.FactorPseudo Fdfp-valuea)HeightZone20.3910.001Time0.8983820.396Zone * Time0.7665820.454Tow [zone, time]16.901350.001b)LengthZone21.86810.001Time0.6773120.461Zone * Time0.5687320.596Tow [zone, time]20.834350.001c)Adductor muscle diameterZone7.100710.012Time6.880620.008Zone * Time0.3118720.756Tow [zone, time]6.2129340.001d)Ovary areaZone13.91810.001Time2.678520.065Zone * Time0.874120.452Tow [zone, time]5.0324340.001e)Testis areaZone14.06510.002Time8.489220.004Zone * Time0.4914220.648Tow [zone, time]5.1403340.001f)Total gonad areaZone11.62610.001Time3.407120.041Zone * Time0.6488320.534Tow [zone, time]8.1704340.001g)Gonad stageZone7.180210.017Time1.219520.303Zone * Time.0859220.938Tow [zone, time]6.1395340.001References ADDIN EN.REFLIST Althaus, F., Hill, N., Ferrari, R., Edwards, L., Przeslawski, R., Sch?nberg, C.H.L., Stuart-Smith, R., Barrett, N., Edgar, G., Colquhoun, J., Tran, M., Jordan, A., Rees, T., Gowlett-Holmes, K., 2015. A Standardised Vocabulary for Identifying Benthic Biota and Substrata from Underwater Imagery: The CATAMI Classification Scheme. PLoS One 10, e0141039.Brand, A.R., 2016. Scallop ecology: distribution and behaviour, in: Shumway, S.E., Jay Parsons, G. (Eds.), Scallops: Bioogy, Ecology, Aquaculture, and Fisheries, 3rd ed. Elsevier, New York, pp. 469-513.Hawkins, A.D., Pembroke, A.E., Popper, A.N., 2015. Information gaps in understanding the effects of noise on fishes and invertebrates. Reviews in Fish Biology and Fisheries 25, 39-64.LeVeque, R., 2007. Finite Difference Methods for Ordinary and Partial Differential Equations: Steady- state and Time Dependent Problems. Society for Industrial and Applied Mathematics.Przeslawski, R., Hurt, L., Forrest, A., Carroll, A., 2016. Potential short-term impacts of marine seismic surveys on scallops in the Gippsland Basin. Fisheries Research and Development Corporation, Canberra, p. 60.Williams, S.B., Pizarro, O.R., Jakuba, M.V., Johnson, C.R., Barrett, N.S., Babcock, R.C., Kendrick, G.A., Steinberg, P.D., Heyward, A.J., Doherty, P.J., 2012. Monitoring of benthic reference sites: using an autonomous underwater vehicle. IEEE Robotics & Automation Magazine 19, 73-84. ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- 2018 a level results sri lanka
- gce a l results 2017
- take five lottery results for tonight
- ny lottery results for friday
- nature methods guide for author
- ny lottery results for sunday
- ny lottery results for saturday
- 2019 a l results check
- lottery numbers results for wednesday
- equation for sound wavelength
- quest lab results for physicians
- blood test results for lupus