This article was downloaded by: [Memorial University of Newfoundland] On: 25 January 2015, At: 07:31 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Biofouling: The Journal of Bioadhesion and Biofilm Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/gbif20
Fouling in your own nest: vessel noise increases biofouling a
b
Jenni A. Stanley , Serena L. Wilkens & Andrew G. Jeffs
a
a
Leigh Marine Laboratory, Institute of Marine Science, University of Auckland, Auckland, New Zealand b
National Institute of Water and Atmospheric Research, Wellington, New Zealand Published online: 07 Aug 2014.
Click for updates To cite this article: Jenni A. Stanley, Serena L. Wilkens & Andrew G. Jeffs (2014) Fouling in your own nest: vessel noise increases biofouling, Biofouling: The Journal of Bioadhesion and Biofilm Research, 30:7, 837-844, DOI: 10.1080/08927014.2014.938062 To link to this article: http://dx.doi.org/10.1080/08927014.2014.938062
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Biofouling, 2014 Vol. 30, No. 7, 837–844, http://dx.doi.org/10.1080/08927014.2014.938062
Fouling in your own nest: vessel noise increases biofouling Jenni A. Stanleya*, Serena L. Wilkensb and Andrew G. Jeffsa a Leigh Marine Laboratory, Institute of Marine Science, University of Auckland, Auckland, New Zealand; bNational Institute of Water and Atmospheric Research, Wellington, New Zealand
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
(Received 19 February 2014; accepted 19 June 2014) Globally billions of dollars are spent each year on attempting to reduce marine biofouling on commercial vessels, largely because it results in higher fuel costs due to increased hydrodynamic drag. Biofouling has been long assumed to be primarily due to the availability of vacant space on the surface of the hull. Here, it is shown that the addition of the noise emitted through a vessel’s hull in port increases the settlement and growth of biofouling organisms within four weeks of clean surfaces being placed in the sea. More than twice as many bryozoans, oysters, calcareous tube worms and barnacles settled and established on surfaces with vessel noise compared to those without. Likewise, individuals from three species grew significantly larger in size in the presence of vessel noise. The results demonstrate that vessel noise in port is promoting biofouling on hulls and that underwater sound plays a much wider ecological role in the marine environment than was previously considered possible. Keywords: Biofouling; vessel noise; bioinvasion; settlement; growth
Introduction The shipping industry is endlessly plagued by marine biofouling, which is the result of the rapid settlement and growth of algae and invertebrates on the submerged surfaces of vessels (Callow & Callow 2011). Biofouling on the hulls of vessels increases the hydrodynamic drag of the vessel, which in turn decreases the top speed and increases fuel consumption (Flemming 2002; Schultz et al. 2011). Many millions of dollars are spent each year on increased fuel costs and in attempting to control biofouling on commercial vessels in the US alone (Schultz et al. 2011). For example, the overall costs associated with managing hull fouling for the US Navy are estimated to be ~ $1 billion for the entire fleet of DDG-51 class destroyers over a period of 15 years (Schultz et al. 2011). Vessel biofouling also poses a significant risk for marine biosecurity owing to the vessel-mediated spread of invasive species in the marine environment (Campbell & Hewitt 2011). For example, in New Zealand, one of the most isolated island groups in the world, it is estimated that more than two thirds of the introduced marine species arrived on vessel hulls (Cranfield et al. 1998). The introduction of nonindigenous species is now acknowledged as a major threat to marine biodiversity and a contributor to marine environmental change (Bax et al. 2003). A wide range of marine taxa foul the hull surfaces of vessels, including micro and macroalgae, barnacles,
sea squirts, and bryozoans (Carlton 1985; Wahl 1989). These groups of organisms are usually characterized by having larvae or propagules that are capable of settling opportunistically and establishing on the vacant or biofilm-coated hard surface of the vessel hull (Cao et al. 2011). There is a large amount of literature that describes the settlement processes and preferences of benthic organisms that are sedentary or sessile as adults, with emphasis placed on biofouling species, as they often degrade commercially important structures. These studies indicate that settlement is a complex process that is regulated by physical and/or chemical cues (for reviews, see Rodriguez et al. 1993; Michael & Valerie 2001; Jackson et al. 2002; Dobretsov et al. 2013; Mieszkin et al. 2013). Chemical and physical cues can include, but are not limited to, inorganic and organic compounds from conspecifics or prey species, biofilms, surface energy, vibration, sound, light, and space availability. For example, cyprid stage barnacles exhibit settlement behavior in response to physical cues such as water flow, pressure, light, surface texture, and chemical cues such as those associated with larval and adult conspecifics and biofilms (Pechenik et al. 1993; Dobretsov et al. 2013). The movement of vessels is also thought to promote the growth of some hull fouling organisms through improved opportunities for feeding (Lehaitre & Compere 2007). Various
*Corresponding author. Email: [email protected] Author contributions: J.S. co-conceived and co-designed the study, undertook the fieldwork, data analysis, and wrote the manuscript. A.J. co-conceived and co-designed the study, and S.W. co-conceived the study. All the authors commented on and provided significant input to the final manuscript. © 2014 Taylor & Francis
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
838
J.A. Stanley et al.
methods of reducing biofouling on ships have been used, mostly involving the use of toxic coatings on vessel hulls to kill or deter the settlement of biofouling organisms (Schultz et al. 2011). However, the widespread use of toxic antifouling chemicals, such as copper and tin compounds, has led to restrictions on their use owing to collateral environmental impacts (Champ 2000). Recent research has determined that natural ambient underwater sound plays an important ecological role in modulating settlement in a number of marine organisms (Montgomery et al. 2006), including some species of fish and invertebrates (Stobutzki & Bellwood 1998; Tolimieri et al. 2000; Leis & Lockett 2005; Simpson et al. 2005; Stanley et al. 2012). For example, the ambient underwater sound emanating from suitable habitats both attracts and promotes faster settlement in the swimming larvae of several crab species (Radford et al. 2007; Stanley et al. 2010, 2012). Over the last 50 years, the growing number of vessel movements has contributed to a 32-fold increase in the low-frequency noise present in some parts of the ocean (Malakoff 2010). Vessels are major sources of underwater noise, especially in the lowfrequency range between 10 and 1,000 Hz (Gotz et al. 2009). Vessel noise is generated by the operation of heavy engines, noisy propellers (cavitation), gears, and large auxiliary engines, such as diesel generators, many of which transmit their vibration energy into the sea, especially via steel hulls (Ross 1976). In the past, the focus of research into the effects of vessel noise in the marine environment has been mostly on its direct negative impacts on certain vertebrate species, especially marine mammals and fish (Nowacek et al. 2001; Weilgart 2007; Codarin et al. 2009). However, the results of recent laboratory research have suggested that the underwater noise generated by vessels in port could be increasing the settlement rate of a few key species of fouling organisms (ie mussels and ascidians) via acoustic settlement cues (Wilkens et al. 2012; McDonald et al. 2014). These previous results have led the present authors to further investigate whether the noise emitted from a vessel in port could have a much wider role in attracting and promoting the settlement of a broad range of important fouling taxa. For this current research, fieldbased settlement assays were employed in conjunction with underwater loud speakers to simulate more closely the acoustic and settlement conditions that are likely to be encountered by fouling larvae in relation to a vessel hull in situ. The results may have important implications for improving the management of vessel fouling and could potentially lead to a reduction in the vesselmediated spread of invasive species. The results may also improve the understanding of the role of underwater sound in the settlement of larvae of marine taxa.
Materials and methods The study was undertaken during April–May 2011 in the temperate waters of Bon Accord Harbor in northern New Zealand (36°25′31.42″S, 174°50′58.02″E) where the biodiversity, abundance, and size of marine biofouling organisms that accumulated on experimental panels in conjunction with vessel noise over 27 days were compared. The timing of the experiment was intended to provide sufficient time for any available fouling larvae to settle and become established to a size at which they could be reliably identified. Underwater sound recording Vessel generator noise was recorded from the MV Straitsman, a 125 m long, 4,168 tonne, steel-hulled, car and passenger ferry whilst at berth with generator engines running in the Port of Wellington, New Zealand. A calibrated hydrophone (High Tech Inc., Mississippi, USA, HTI-96-Min: sensitivity –164.3 dB re 1 V/μPa; frequency range 10 Hz to 30 kHz flat response) was used to record a series of 5 min samples of underwater noise emitted by the vessel. The vessel was operating a shipbased generator power supply, typical for vessels at berth, and the main forward propulsion engines were off. No other machinery was in operation during the recordings. There is variability in the sound emitted by individual vessels. However, it is largely in response to characteristics such as size (increasing size decreases frequency), hull type, and the operation of additional machinery on board (Hildebrand 2009). The underwater acoustic output emitted from the hull of the MV Straitsman is an accurate representation of vessels >100 m in hull length that are usually used for international trade, as its acoustic characteristics are similar to other steelhulled vessels of this size whilst at berth (personal observations). The hydrophone was placed ~1 m from the hull, port side at midship (the approximate location of the generator on this vessel) and lowered 2 m into the water. The recording was captured using a calibrated digital recorder (Edirol R09HR 24-bit recorder; sampling rate 48 kHz, Roland Corporation, Japan). Digital recordings were downloaded to a PC, analyzed, and the spectral composition and sound source level determined using MATLAB software (MathWorks, Inc.) with codes written specifically for the analysis of underwater sound recording. Power spectra were generated using a fast Fourier transformation analysis of eight 10 s samples, which were randomly selected from the 5 min recordings and smoothed using an 11-point triangular window. Sound pressure levels between 30 and 10,000 Hz were determined from each of the eight 10 s samples and the means used for the acoustic levels used in the experimental treatments.
Biofouling
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Experimental sound source The field-based experiment consisted of six underwater speakers (Lubell Labs Inc., Columbus, OH; LL964, frequency response 0.2–20 kHz, omnidirectional sound transmission), three of which were playing vessel noise (Sound treatment), and three were controls with no replayed sound (Silent treatment). The underwater speakers in the Sound treatment continuously broadcast vessel noise at 128 dB re 1 μPa RMS level in the 30–10,000 Hz range, and the Silent treatment remained silent. The underwater speakers were driven by efficient piezoelectric drivers with the pressure shells composed of inert epoxy-coated aluminum shells (lubell.com) and therefore unlikely to produce any magnetic or electric fields that might influence larval behavior in the vicinity. The underwater speakers were deployed by suspending them in ~3 m of water depth at low tide, ~1.5 m from the seabed beneath a 0.5 km long decommissioned wharf in Bon Accord Harbor. The Sound and Silent treatments were intermixed along either side of the length of the wharf to reduce acoustic overlap as much as possible between the two treatments while still sampling the same larval supply. A Sony MP3 player and amplifier was used to power each of the underwater speakers in the Sound treatment that continuously replayed a 4 min sequence of pre-recorded vessel noise. A calibrated hydrophone and recorder (High Tech, Inc., Mississippi, USA HTI-96-MIN, Sound Devices, WI, 722 recorder) was used to confirm the speaker produced the same sound source level as measured from the vessel at the time of the recordings. Different 4 min sequences of pre-recorded sound were used for each of the three underwater speakers to avoid any pseudoreplication of the initial sound treatment. The three underwater speakers in the Silent treatment did not replay any prerecorded sound to act as a control for any attraction of fouling organisms to these objects. Experimental panels For each underwater speaker, a flat fiber-cement panel (200 × 200 mm) was attached to each of the opposing surfaces of the speaker, one facing downward and the other upward. The experimental panels consisted of HardiflexTM (fiber-cement board) that had been presoaked for 7 days in filtered seawater (1 μm) prior to the experiments commencing to allow the formation of bacterial biofilms, thus enhancing the settlement surface for the experiment. The 1 μm filter was used to prevent any marine larval invertebrates settling on the plates prior to deployment in the field. Presoaking was also used to remove any possible residual soluble chemicals from manufacturing that may have interfered with larval settlement (Balch & Scheibling 2000). The experimental
839
panels were examined under a microscope (40 × magnification) to ensure no macroscopic fouling organisms had settled prior to the attachment of the panels to the underwater speakers. The presoaked settlement panels were attached directly to the flat opposing sides of the underwater speakers using waterproof silicone glue to ensure the panel moved synchronously with the speaker and held in place with elastic cords (not covering any part of the panel surface); therefore, the Sound and Silent treatments each comprised three downward facing and three upward facing panels. At the conclusion of the experiment, divers collected the speakers, and the experimental panels were removed and carefully placed in individual sealed plastic bags to reduce loss of any organisms when transporting to the laboratory for analysis. Analysis of the fouling on the experimental panels involved counting and identifying to species level where possible all sessile fouling organisms with a 40× magnification dissecting microscope. The size of the organisms was measured using digital calipers: maximum shell diameter for oysters and barnacles, maximum colony diameter for encrusting bryozoans, and stalk length for branching bryozoan. The sizes of the calcareous tube worms and barnacle cyprids were not measured. Statistical analyses A two-way ANOVA tested whether the mean total number of fouling organisms on an experimental panel was affected by treatment (Sound vs Silent; fixed factor) and orientation (downward and upward oriented; fixed factor). A priori the normality of the data was confirmed using the Kolmogorov–Smirnov test (p = 0.084) and homogeneity of variances confirmed with Levene’s Median test. A t-test was used to detect differences in the mean number of individuals of each species per plate for Sound vs Silent treatments for both upward and downward oriented panels. Mann–Whitney Rank Sum tests were used to test for differences in the median size of individuals within a species for Sound vs Silent treatments. The ordination technique non-metric multidimensional scaling (MDS) was used to examine relationships between the mean number of individuals in each treatment and orientation. MDS creates low-dimensional maps of relationships among treatments and orientation, where the distance between points is proportional to their multivariate similarity. The analysis was run on a Bray– Curtis dissimilarity matrix derived from forth-root transformed density data. Analysis of similarity (ANOSIM) was then used to test for differences among these treatments and orientations. All analyses were run using Primer V6.1.12 (PRIMER-E, Plymouth, UK) and SigmaPlot 11 (Systat Software Inc.).
840
J.A. Stanley et al.
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Results Underwater sound recording and experimental sound source The vessel noise sound pressure level was measured at 128 dB re 1 μPa RMS in the 30–10,000 Hz range, 10 s sample, at the source (1 m) and was largely dominated by low frequencies, 30 – 2,000 Hz (Figure 1). The sound broadcast in the Sound treatments had a similar overall spectral composition and intensity to the source signals recorded from the vessel at berth, with low frequencies in the range of 30–2,000 Hz dominating (Figure 1). The Silent treatments had little sound transfer from the Sound treatments with mostly ambient underwater sounds from the harbor present (Figure 1). Experimental panels Initial visual inspection of the Sound vs Silent experimental panels and speaker housings after 27 days in the water revealed much greater fouling by Bugula neritina and Pomatoceros sp. on those in the Sound treatment (Figure 2). It also revealed differences in the total abundance of fouling organisms between the upward and downward oriented experimental panels within and among treatments.
Figure 1. Power spectra of 10 s samples showing the spectral composition and sound pressure level of recorded vessel noise, broadcast vessel noise (Sound treatment) and ambient underwater sound (Silent treatment) at Moores Bay, Kawau Island, northern New Zealand. The black line represents the original vessel hull noise when recorded in port, the grey line represents vessel noise when broadcast from underwater speaker when recorded from Sound treatment and the light grey line represents ambient underwater sound when recorded from the Silent treatment.
Figure 2. Photographs showing the visual difference in biofouling of Sound vs Silent treatment. (A) Silent treatment speaker housing, (B) Sound treatment speaker housing, (C) example of Silent experimental panel, downward orientation and, D), example of Sound experimental panel, downward orientation.
The total number of organisms found on the downward- and upward-oriented panels was consistently around three times higher in the Sound treatment than in the Silent treatment, ie, total number of individuals; 2,190 vs 757 for downward oriented and 397 vs 120 for upward oriented. There was a statistically significant difference between treatments for total organism count on the experimental panels (FANOVA = 142.8, PANOVA = < 0.001) and panel orientation (F = 288.4, p < 0.001) and the interaction between treatment and panel orientation (F = 65.3, p < 0.001). A total of eight fouling species was found attached to the experimental panels, viz. bryozoans, Bugula neritina (erect branching), Watersipora arcuata (encrusting), Watersipora subtorquata (encrusting), oysters, Crassostrea gigas, Ostrea chilensis, calcareous tube worm, Pomatoceros sp., barnacles, Elminius modestus, Balanus amphitrite, and also barnacle cyprids that could not be identified to species. On the downward-oriented experimental panels, the mean number of individuals of six of the eight species, as well as the barnacle cyprids, were consistently higher on panels in the Sound vs the Silent treatment; Pomatoceros sp. (t = 3.21, p = 0.030), W. arcuata (t = 3.01, p = 0.040), B. neritina (tt-test = 4.85, P t-test = 0.008), E. modestus (t = 8.21, p = 0.001), C. gigas (t = 5.60, p = 0.005), O. chilensis (t = 6.27, p = 0.003), barnacle cyprids (t = 12.02, p < 0.001) (Figure 3A). However, there was no significant difference in the mean numbers of individuals of W. subtorquata and B. amphitrite,
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Biofouling
841
Figure 4. Non-metric multidimensional scaling (MDS) analysis of the total number of organisms attached to experimental panels in Sound and Silent treatments and with downward and upward orientations of experimental panels.
distinctions from each other (RANOSIM = 0.796, p = 0.001) (Figure 4). In addition to a significantly higher number of individuals, the median sizes of individuals of four of the seven species that were measured for size were significantly larger in the Sound than in the Silent treatments: W. arcuata (Sound 4 mm vs Silent 3.2 mm, percentiles Sound 1.9 and 6.0, Silent 1.9 and 4.8; UMann–Whitney = 27952.5, PMann–Whitney = 0.004), B. neritina (Sound 6 mm vs Silent 3.9 mm, percentiles Sound 3.4 and 9.1, Silent Figure 3. The mean (+SE) number of individuals of each species per experimental panel for A, downward oriented; Sound treatments (black), Silent treatments (grey) and B, upward oriented; Sound treatment (black), Silent treatment (grey). Statistical results for t-tests, * < 0.05, ** < 0.01, *** ≤ 0.001.
which was due to very low abundances found in both treatments. On the upward-oriented experimental panels, all eight species and barnacle cyprids were present in the Sound treatments. However, the barnacle B. amphitrite and the unidentifiable cyprids were absent in the Silent treatments. Therefore, the mean numbers of individuals of six species and barnacle cyprids were consistently higher on the experimental panels in the Sound than in the Silent treatment: B. neritina (tt-test = 4.06, Pt-test = 0.015), W. arcuata (t = 4.20, p = 0.014), Pomatoceros sp. (t = 4.42, p = 0.011), E. modestus (t = 6.89, p = 0.002), C. gigas (t = 5.56, p = 0.003), and B. amphitrite (Figure 3B). The mean number of individuals was clearly grouped in multivariate space according to treatment and treatment × orientation (Figure 5). All treatments and experimental panel orientations showed strong grouping
Figure 5. Size frequency of all the erect bryozoan, Bugula neritina, individuals found on Sound and Silent treatment settlement panels (n = 6 panels for each treatment). The black bars represent Sound treatments and the grey bars represent Silent treatments.
842
J.A. Stanley et al.
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
2.8, 5.6; U = 19441.0, p 2 mm) on the surface of the upward-oriented panels and not on the downwardoriented panels. Small prostrate settlers could be easily smothered by the settling silt, and for most settling larvae, even a thin coating of silt would make attachment directly to the cement panel difficult or impossible, depending on the degree of sedimentation (Irving & Connell 2002). The results of the current study provide the first evidence from an experiment conducted in natural conditions that the underwater sound emitted by vessels in port is attracting the larvae and promoting their settlement, and increasing growth of a number of globally important fouling species. It is likely that the larvae of these fouling organisms are being attracted to the vessel hulls as a result of the noise. This reasoning is consistent with previous studies that have shown that the settlement-stage larvae of several marine invertebrates, such as crabs and coral, are actively swimming towards the natural sounds emanating from their preferred settlement habitats (Radford et al. 2007; Vermeij et al. 2010). It has also been shown that the speed of settlement, metamorphosis, and subsequent moulting of decapod crustaceans can be substantially increased when exposed to the ambient underwater sound associated with their preferred settlement habitats (Stanley et al. 2012). The same phenomenon has also been observed in experiments in the laboratory where a species of a bivalve mollusc and a solitary ascidian were exposed to the recorded ambient noise emitted from a steel-hulled vessel in port (Wilkens et al. 2012; McDonald et al. 2014). The current results and those mentioned above are in contrast to the few previous investigations into the responses of barnacles to acoustic and vibrational stimuli. Branscomb and Rittschof (1984) observed that lowfrequency sound, 15–45 Hz, inhibited attachment and metamorphosis of laboratory-reared and plankton-caught barnacle cyprids. Furthermore, Guo et al. (2011, 2012) observed that ultrasound (23, 63, and 102 kHz) lowered the settlement rates and increased mortality in laboratoryreared cyprids of Amphibalanus amphitrite, probably as a result of increased cavitation around the settled barnacle (Guo et al. 2011). The settlement of the larvae of two barnacle species was found to be inhibited at vibration
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Biofouling frequencies > 260 Hz when vibrational excitation of a range of frequencies and amplitudes was applied to experimental fouling panels, whereas there was no effect on all other fouling species, including tubeworms, bryozoans, ascidians, and algae (Choi et al. 2013). These contrasting results suggest that it may not be the dominant low frequencies (
Biofouling: The Journal of Bioadhesion and Biofilm Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/gbif20
Fouling in your own nest: vessel noise increases biofouling a
b
Jenni A. Stanley , Serena L. Wilkens & Andrew G. Jeffs
a
a
Leigh Marine Laboratory, Institute of Marine Science, University of Auckland, Auckland, New Zealand b
National Institute of Water and Atmospheric Research, Wellington, New Zealand Published online: 07 Aug 2014.
Click for updates To cite this article: Jenni A. Stanley, Serena L. Wilkens & Andrew G. Jeffs (2014) Fouling in your own nest: vessel noise increases biofouling, Biofouling: The Journal of Bioadhesion and Biofilm Research, 30:7, 837-844, DOI: 10.1080/08927014.2014.938062 To link to this article: http://dx.doi.org/10.1080/08927014.2014.938062
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Biofouling, 2014 Vol. 30, No. 7, 837–844, http://dx.doi.org/10.1080/08927014.2014.938062
Fouling in your own nest: vessel noise increases biofouling Jenni A. Stanleya*, Serena L. Wilkensb and Andrew G. Jeffsa a Leigh Marine Laboratory, Institute of Marine Science, University of Auckland, Auckland, New Zealand; bNational Institute of Water and Atmospheric Research, Wellington, New Zealand
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
(Received 19 February 2014; accepted 19 June 2014) Globally billions of dollars are spent each year on attempting to reduce marine biofouling on commercial vessels, largely because it results in higher fuel costs due to increased hydrodynamic drag. Biofouling has been long assumed to be primarily due to the availability of vacant space on the surface of the hull. Here, it is shown that the addition of the noise emitted through a vessel’s hull in port increases the settlement and growth of biofouling organisms within four weeks of clean surfaces being placed in the sea. More than twice as many bryozoans, oysters, calcareous tube worms and barnacles settled and established on surfaces with vessel noise compared to those without. Likewise, individuals from three species grew significantly larger in size in the presence of vessel noise. The results demonstrate that vessel noise in port is promoting biofouling on hulls and that underwater sound plays a much wider ecological role in the marine environment than was previously considered possible. Keywords: Biofouling; vessel noise; bioinvasion; settlement; growth
Introduction The shipping industry is endlessly plagued by marine biofouling, which is the result of the rapid settlement and growth of algae and invertebrates on the submerged surfaces of vessels (Callow & Callow 2011). Biofouling on the hulls of vessels increases the hydrodynamic drag of the vessel, which in turn decreases the top speed and increases fuel consumption (Flemming 2002; Schultz et al. 2011). Many millions of dollars are spent each year on increased fuel costs and in attempting to control biofouling on commercial vessels in the US alone (Schultz et al. 2011). For example, the overall costs associated with managing hull fouling for the US Navy are estimated to be ~ $1 billion for the entire fleet of DDG-51 class destroyers over a period of 15 years (Schultz et al. 2011). Vessel biofouling also poses a significant risk for marine biosecurity owing to the vessel-mediated spread of invasive species in the marine environment (Campbell & Hewitt 2011). For example, in New Zealand, one of the most isolated island groups in the world, it is estimated that more than two thirds of the introduced marine species arrived on vessel hulls (Cranfield et al. 1998). The introduction of nonindigenous species is now acknowledged as a major threat to marine biodiversity and a contributor to marine environmental change (Bax et al. 2003). A wide range of marine taxa foul the hull surfaces of vessels, including micro and macroalgae, barnacles,
sea squirts, and bryozoans (Carlton 1985; Wahl 1989). These groups of organisms are usually characterized by having larvae or propagules that are capable of settling opportunistically and establishing on the vacant or biofilm-coated hard surface of the vessel hull (Cao et al. 2011). There is a large amount of literature that describes the settlement processes and preferences of benthic organisms that are sedentary or sessile as adults, with emphasis placed on biofouling species, as they often degrade commercially important structures. These studies indicate that settlement is a complex process that is regulated by physical and/or chemical cues (for reviews, see Rodriguez et al. 1993; Michael & Valerie 2001; Jackson et al. 2002; Dobretsov et al. 2013; Mieszkin et al. 2013). Chemical and physical cues can include, but are not limited to, inorganic and organic compounds from conspecifics or prey species, biofilms, surface energy, vibration, sound, light, and space availability. For example, cyprid stage barnacles exhibit settlement behavior in response to physical cues such as water flow, pressure, light, surface texture, and chemical cues such as those associated with larval and adult conspecifics and biofilms (Pechenik et al. 1993; Dobretsov et al. 2013). The movement of vessels is also thought to promote the growth of some hull fouling organisms through improved opportunities for feeding (Lehaitre & Compere 2007). Various
*Corresponding author. Email: [email protected] Author contributions: J.S. co-conceived and co-designed the study, undertook the fieldwork, data analysis, and wrote the manuscript. A.J. co-conceived and co-designed the study, and S.W. co-conceived the study. All the authors commented on and provided significant input to the final manuscript. © 2014 Taylor & Francis
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
838
J.A. Stanley et al.
methods of reducing biofouling on ships have been used, mostly involving the use of toxic coatings on vessel hulls to kill or deter the settlement of biofouling organisms (Schultz et al. 2011). However, the widespread use of toxic antifouling chemicals, such as copper and tin compounds, has led to restrictions on their use owing to collateral environmental impacts (Champ 2000). Recent research has determined that natural ambient underwater sound plays an important ecological role in modulating settlement in a number of marine organisms (Montgomery et al. 2006), including some species of fish and invertebrates (Stobutzki & Bellwood 1998; Tolimieri et al. 2000; Leis & Lockett 2005; Simpson et al. 2005; Stanley et al. 2012). For example, the ambient underwater sound emanating from suitable habitats both attracts and promotes faster settlement in the swimming larvae of several crab species (Radford et al. 2007; Stanley et al. 2010, 2012). Over the last 50 years, the growing number of vessel movements has contributed to a 32-fold increase in the low-frequency noise present in some parts of the ocean (Malakoff 2010). Vessels are major sources of underwater noise, especially in the lowfrequency range between 10 and 1,000 Hz (Gotz et al. 2009). Vessel noise is generated by the operation of heavy engines, noisy propellers (cavitation), gears, and large auxiliary engines, such as diesel generators, many of which transmit their vibration energy into the sea, especially via steel hulls (Ross 1976). In the past, the focus of research into the effects of vessel noise in the marine environment has been mostly on its direct negative impacts on certain vertebrate species, especially marine mammals and fish (Nowacek et al. 2001; Weilgart 2007; Codarin et al. 2009). However, the results of recent laboratory research have suggested that the underwater noise generated by vessels in port could be increasing the settlement rate of a few key species of fouling organisms (ie mussels and ascidians) via acoustic settlement cues (Wilkens et al. 2012; McDonald et al. 2014). These previous results have led the present authors to further investigate whether the noise emitted from a vessel in port could have a much wider role in attracting and promoting the settlement of a broad range of important fouling taxa. For this current research, fieldbased settlement assays were employed in conjunction with underwater loud speakers to simulate more closely the acoustic and settlement conditions that are likely to be encountered by fouling larvae in relation to a vessel hull in situ. The results may have important implications for improving the management of vessel fouling and could potentially lead to a reduction in the vesselmediated spread of invasive species. The results may also improve the understanding of the role of underwater sound in the settlement of larvae of marine taxa.
Materials and methods The study was undertaken during April–May 2011 in the temperate waters of Bon Accord Harbor in northern New Zealand (36°25′31.42″S, 174°50′58.02″E) where the biodiversity, abundance, and size of marine biofouling organisms that accumulated on experimental panels in conjunction with vessel noise over 27 days were compared. The timing of the experiment was intended to provide sufficient time for any available fouling larvae to settle and become established to a size at which they could be reliably identified. Underwater sound recording Vessel generator noise was recorded from the MV Straitsman, a 125 m long, 4,168 tonne, steel-hulled, car and passenger ferry whilst at berth with generator engines running in the Port of Wellington, New Zealand. A calibrated hydrophone (High Tech Inc., Mississippi, USA, HTI-96-Min: sensitivity –164.3 dB re 1 V/μPa; frequency range 10 Hz to 30 kHz flat response) was used to record a series of 5 min samples of underwater noise emitted by the vessel. The vessel was operating a shipbased generator power supply, typical for vessels at berth, and the main forward propulsion engines were off. No other machinery was in operation during the recordings. There is variability in the sound emitted by individual vessels. However, it is largely in response to characteristics such as size (increasing size decreases frequency), hull type, and the operation of additional machinery on board (Hildebrand 2009). The underwater acoustic output emitted from the hull of the MV Straitsman is an accurate representation of vessels >100 m in hull length that are usually used for international trade, as its acoustic characteristics are similar to other steelhulled vessels of this size whilst at berth (personal observations). The hydrophone was placed ~1 m from the hull, port side at midship (the approximate location of the generator on this vessel) and lowered 2 m into the water. The recording was captured using a calibrated digital recorder (Edirol R09HR 24-bit recorder; sampling rate 48 kHz, Roland Corporation, Japan). Digital recordings were downloaded to a PC, analyzed, and the spectral composition and sound source level determined using MATLAB software (MathWorks, Inc.) with codes written specifically for the analysis of underwater sound recording. Power spectra were generated using a fast Fourier transformation analysis of eight 10 s samples, which were randomly selected from the 5 min recordings and smoothed using an 11-point triangular window. Sound pressure levels between 30 and 10,000 Hz were determined from each of the eight 10 s samples and the means used for the acoustic levels used in the experimental treatments.
Biofouling
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Experimental sound source The field-based experiment consisted of six underwater speakers (Lubell Labs Inc., Columbus, OH; LL964, frequency response 0.2–20 kHz, omnidirectional sound transmission), three of which were playing vessel noise (Sound treatment), and three were controls with no replayed sound (Silent treatment). The underwater speakers in the Sound treatment continuously broadcast vessel noise at 128 dB re 1 μPa RMS level in the 30–10,000 Hz range, and the Silent treatment remained silent. The underwater speakers were driven by efficient piezoelectric drivers with the pressure shells composed of inert epoxy-coated aluminum shells (lubell.com) and therefore unlikely to produce any magnetic or electric fields that might influence larval behavior in the vicinity. The underwater speakers were deployed by suspending them in ~3 m of water depth at low tide, ~1.5 m from the seabed beneath a 0.5 km long decommissioned wharf in Bon Accord Harbor. The Sound and Silent treatments were intermixed along either side of the length of the wharf to reduce acoustic overlap as much as possible between the two treatments while still sampling the same larval supply. A Sony MP3 player and amplifier was used to power each of the underwater speakers in the Sound treatment that continuously replayed a 4 min sequence of pre-recorded vessel noise. A calibrated hydrophone and recorder (High Tech, Inc., Mississippi, USA HTI-96-MIN, Sound Devices, WI, 722 recorder) was used to confirm the speaker produced the same sound source level as measured from the vessel at the time of the recordings. Different 4 min sequences of pre-recorded sound were used for each of the three underwater speakers to avoid any pseudoreplication of the initial sound treatment. The three underwater speakers in the Silent treatment did not replay any prerecorded sound to act as a control for any attraction of fouling organisms to these objects. Experimental panels For each underwater speaker, a flat fiber-cement panel (200 × 200 mm) was attached to each of the opposing surfaces of the speaker, one facing downward and the other upward. The experimental panels consisted of HardiflexTM (fiber-cement board) that had been presoaked for 7 days in filtered seawater (1 μm) prior to the experiments commencing to allow the formation of bacterial biofilms, thus enhancing the settlement surface for the experiment. The 1 μm filter was used to prevent any marine larval invertebrates settling on the plates prior to deployment in the field. Presoaking was also used to remove any possible residual soluble chemicals from manufacturing that may have interfered with larval settlement (Balch & Scheibling 2000). The experimental
839
panels were examined under a microscope (40 × magnification) to ensure no macroscopic fouling organisms had settled prior to the attachment of the panels to the underwater speakers. The presoaked settlement panels were attached directly to the flat opposing sides of the underwater speakers using waterproof silicone glue to ensure the panel moved synchronously with the speaker and held in place with elastic cords (not covering any part of the panel surface); therefore, the Sound and Silent treatments each comprised three downward facing and three upward facing panels. At the conclusion of the experiment, divers collected the speakers, and the experimental panels were removed and carefully placed in individual sealed plastic bags to reduce loss of any organisms when transporting to the laboratory for analysis. Analysis of the fouling on the experimental panels involved counting and identifying to species level where possible all sessile fouling organisms with a 40× magnification dissecting microscope. The size of the organisms was measured using digital calipers: maximum shell diameter for oysters and barnacles, maximum colony diameter for encrusting bryozoans, and stalk length for branching bryozoan. The sizes of the calcareous tube worms and barnacle cyprids were not measured. Statistical analyses A two-way ANOVA tested whether the mean total number of fouling organisms on an experimental panel was affected by treatment (Sound vs Silent; fixed factor) and orientation (downward and upward oriented; fixed factor). A priori the normality of the data was confirmed using the Kolmogorov–Smirnov test (p = 0.084) and homogeneity of variances confirmed with Levene’s Median test. A t-test was used to detect differences in the mean number of individuals of each species per plate for Sound vs Silent treatments for both upward and downward oriented panels. Mann–Whitney Rank Sum tests were used to test for differences in the median size of individuals within a species for Sound vs Silent treatments. The ordination technique non-metric multidimensional scaling (MDS) was used to examine relationships between the mean number of individuals in each treatment and orientation. MDS creates low-dimensional maps of relationships among treatments and orientation, where the distance between points is proportional to their multivariate similarity. The analysis was run on a Bray– Curtis dissimilarity matrix derived from forth-root transformed density data. Analysis of similarity (ANOSIM) was then used to test for differences among these treatments and orientations. All analyses were run using Primer V6.1.12 (PRIMER-E, Plymouth, UK) and SigmaPlot 11 (Systat Software Inc.).
840
J.A. Stanley et al.
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Results Underwater sound recording and experimental sound source The vessel noise sound pressure level was measured at 128 dB re 1 μPa RMS in the 30–10,000 Hz range, 10 s sample, at the source (1 m) and was largely dominated by low frequencies, 30 – 2,000 Hz (Figure 1). The sound broadcast in the Sound treatments had a similar overall spectral composition and intensity to the source signals recorded from the vessel at berth, with low frequencies in the range of 30–2,000 Hz dominating (Figure 1). The Silent treatments had little sound transfer from the Sound treatments with mostly ambient underwater sounds from the harbor present (Figure 1). Experimental panels Initial visual inspection of the Sound vs Silent experimental panels and speaker housings after 27 days in the water revealed much greater fouling by Bugula neritina and Pomatoceros sp. on those in the Sound treatment (Figure 2). It also revealed differences in the total abundance of fouling organisms between the upward and downward oriented experimental panels within and among treatments.
Figure 1. Power spectra of 10 s samples showing the spectral composition and sound pressure level of recorded vessel noise, broadcast vessel noise (Sound treatment) and ambient underwater sound (Silent treatment) at Moores Bay, Kawau Island, northern New Zealand. The black line represents the original vessel hull noise when recorded in port, the grey line represents vessel noise when broadcast from underwater speaker when recorded from Sound treatment and the light grey line represents ambient underwater sound when recorded from the Silent treatment.
Figure 2. Photographs showing the visual difference in biofouling of Sound vs Silent treatment. (A) Silent treatment speaker housing, (B) Sound treatment speaker housing, (C) example of Silent experimental panel, downward orientation and, D), example of Sound experimental panel, downward orientation.
The total number of organisms found on the downward- and upward-oriented panels was consistently around three times higher in the Sound treatment than in the Silent treatment, ie, total number of individuals; 2,190 vs 757 for downward oriented and 397 vs 120 for upward oriented. There was a statistically significant difference between treatments for total organism count on the experimental panels (FANOVA = 142.8, PANOVA = < 0.001) and panel orientation (F = 288.4, p < 0.001) and the interaction between treatment and panel orientation (F = 65.3, p < 0.001). A total of eight fouling species was found attached to the experimental panels, viz. bryozoans, Bugula neritina (erect branching), Watersipora arcuata (encrusting), Watersipora subtorquata (encrusting), oysters, Crassostrea gigas, Ostrea chilensis, calcareous tube worm, Pomatoceros sp., barnacles, Elminius modestus, Balanus amphitrite, and also barnacle cyprids that could not be identified to species. On the downward-oriented experimental panels, the mean number of individuals of six of the eight species, as well as the barnacle cyprids, were consistently higher on panels in the Sound vs the Silent treatment; Pomatoceros sp. (t = 3.21, p = 0.030), W. arcuata (t = 3.01, p = 0.040), B. neritina (tt-test = 4.85, P t-test = 0.008), E. modestus (t = 8.21, p = 0.001), C. gigas (t = 5.60, p = 0.005), O. chilensis (t = 6.27, p = 0.003), barnacle cyprids (t = 12.02, p < 0.001) (Figure 3A). However, there was no significant difference in the mean numbers of individuals of W. subtorquata and B. amphitrite,
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Biofouling
841
Figure 4. Non-metric multidimensional scaling (MDS) analysis of the total number of organisms attached to experimental panels in Sound and Silent treatments and with downward and upward orientations of experimental panels.
distinctions from each other (RANOSIM = 0.796, p = 0.001) (Figure 4). In addition to a significantly higher number of individuals, the median sizes of individuals of four of the seven species that were measured for size were significantly larger in the Sound than in the Silent treatments: W. arcuata (Sound 4 mm vs Silent 3.2 mm, percentiles Sound 1.9 and 6.0, Silent 1.9 and 4.8; UMann–Whitney = 27952.5, PMann–Whitney = 0.004), B. neritina (Sound 6 mm vs Silent 3.9 mm, percentiles Sound 3.4 and 9.1, Silent Figure 3. The mean (+SE) number of individuals of each species per experimental panel for A, downward oriented; Sound treatments (black), Silent treatments (grey) and B, upward oriented; Sound treatment (black), Silent treatment (grey). Statistical results for t-tests, * < 0.05, ** < 0.01, *** ≤ 0.001.
which was due to very low abundances found in both treatments. On the upward-oriented experimental panels, all eight species and barnacle cyprids were present in the Sound treatments. However, the barnacle B. amphitrite and the unidentifiable cyprids were absent in the Silent treatments. Therefore, the mean numbers of individuals of six species and barnacle cyprids were consistently higher on the experimental panels in the Sound than in the Silent treatment: B. neritina (tt-test = 4.06, Pt-test = 0.015), W. arcuata (t = 4.20, p = 0.014), Pomatoceros sp. (t = 4.42, p = 0.011), E. modestus (t = 6.89, p = 0.002), C. gigas (t = 5.56, p = 0.003), and B. amphitrite (Figure 3B). The mean number of individuals was clearly grouped in multivariate space according to treatment and treatment × orientation (Figure 5). All treatments and experimental panel orientations showed strong grouping
Figure 5. Size frequency of all the erect bryozoan, Bugula neritina, individuals found on Sound and Silent treatment settlement panels (n = 6 panels for each treatment). The black bars represent Sound treatments and the grey bars represent Silent treatments.
842
J.A. Stanley et al.
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
2.8, 5.6; U = 19441.0, p 2 mm) on the surface of the upward-oriented panels and not on the downwardoriented panels. Small prostrate settlers could be easily smothered by the settling silt, and for most settling larvae, even a thin coating of silt would make attachment directly to the cement panel difficult or impossible, depending on the degree of sedimentation (Irving & Connell 2002). The results of the current study provide the first evidence from an experiment conducted in natural conditions that the underwater sound emitted by vessels in port is attracting the larvae and promoting their settlement, and increasing growth of a number of globally important fouling species. It is likely that the larvae of these fouling organisms are being attracted to the vessel hulls as a result of the noise. This reasoning is consistent with previous studies that have shown that the settlement-stage larvae of several marine invertebrates, such as crabs and coral, are actively swimming towards the natural sounds emanating from their preferred settlement habitats (Radford et al. 2007; Vermeij et al. 2010). It has also been shown that the speed of settlement, metamorphosis, and subsequent moulting of decapod crustaceans can be substantially increased when exposed to the ambient underwater sound associated with their preferred settlement habitats (Stanley et al. 2012). The same phenomenon has also been observed in experiments in the laboratory where a species of a bivalve mollusc and a solitary ascidian were exposed to the recorded ambient noise emitted from a steel-hulled vessel in port (Wilkens et al. 2012; McDonald et al. 2014). The current results and those mentioned above are in contrast to the few previous investigations into the responses of barnacles to acoustic and vibrational stimuli. Branscomb and Rittschof (1984) observed that lowfrequency sound, 15–45 Hz, inhibited attachment and metamorphosis of laboratory-reared and plankton-caught barnacle cyprids. Furthermore, Guo et al. (2011, 2012) observed that ultrasound (23, 63, and 102 kHz) lowered the settlement rates and increased mortality in laboratoryreared cyprids of Amphibalanus amphitrite, probably as a result of increased cavitation around the settled barnacle (Guo et al. 2011). The settlement of the larvae of two barnacle species was found to be inhibited at vibration
Downloaded by [Memorial University of Newfoundland] at 07:31 25 January 2015
Biofouling frequencies > 260 Hz when vibrational excitation of a range of frequencies and amplitudes was applied to experimental fouling panels, whereas there was no effect on all other fouling species, including tubeworms, bryozoans, ascidians, and algae (Choi et al. 2013). These contrasting results suggest that it may not be the dominant low frequencies (
