� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � ! " # " ! $ % � � � � $ & � ' () � � � � � � � � � � � � � * � � � � � �+ � � � � � � � � � , � � � � � � - , � � � � � � � � � � , � � � � � . � � � � $ ! & " ( / % � , � � � � � � � � � � � � � � � � � � 0 � � �' & 1 % $ � � � � � � � 2 � � � � � � � � � $ � ' # 0 � 3 4

5 6 7 8 9 : ; < = > ? @ A B CBiofouling: The Journal of Bioadhesion and Biofilm

ResearchPublication details, including instructions for authors and subscription information:

http://www.tandfonline.com/loi/gbif20

Vessel generator noise as a settlement cue for marine

biofouling speciesJ.I. McDonald

a, S.L. Wilkens

b, J.A. Stanley

c & A.G. Jeffs

c

a Western Australian Department of Fisheries, WA Fisheries and Marine Research

Laboratories, Perth, Australiab National Institute of Water and Atmospheric Research, Marine Biodiversity and Biosecurity,

Kilbirnie, Wellington, New Zealandc University of Auckland, Auckland, New Zealand

Published online: 28 May 2014.

To cite this article: J.I. McDonald, S.L. Wilkens, J.A. Stanley & A.G. Jeffs (2014): Vessel generator noise as a

settlement cue for marine biofouling species, Biofouling: The Journal of Bioadhesion and Biofilm Research, DOI:

10.1080/08927014.2014.919630

To link to this article: http://dx.doi.org/10.1080/08927014.2014.919630) � - � 2 - 2 D � � � � � � � . � � � � � � + D � -� � � � � � * � � � � � � � � E � F � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � G � � H D � � � I J � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � K 0 � F � � � � � � � � * � � � � � � � � � � , � � � � � � � � � � � � � � � � E �� � � � � � � � � � � � � � � � � � � � � � � � F � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � D � � � K � � � � � � � � � � F � � L � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � F � � � � � � � � � � � � � � �� � � � � � F � � � � � � � � � � � � � � � � � � � * � � � � � � K � � � � � � � � � � � � � D � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � F � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � K � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � , � � � � � � � � � � � � � � L � � � � � � � � , � � � � � � � � � � � � � � � � � � � � � � � F �� � � � � � F � � � � � � � � � � , � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � , � � � � � � � � �� � D � � � K� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � , � � � � � � F � � � � � � � � � � � � K � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � , � � � � � � � 1 � � � � � , � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � L � � � � � � � � � � � � � K � � � � * D � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � M M K � � � � � � � K � � � M � � , M � � � � 1 � � 1 � � � � � � � �

Vessel generator noise as a settlement cue for marine biofouling species

J.I. McDonalda, S.L. Wilkensb*, J.A. Stanleyc and A.G. Jeffsc

aWestern Australian Department of Fisheries, WA Fisheries and Marine Research Laboratories, Perth, Australia;bNational Institute of Water and Atmospheric Research, Marine Biodiversity and Biosecurity, Kilbirnie, Wellington, New Zealand;cUniversity of Auckland, Auckland, New Zealand

(Received 17 December 2013; accepted 25 April 2014)

Underwater noise is increasing globally, largely due to increased vessel numbers and international ocean trade. Vesselsare also a major vector for translocation of non-indigenous marine species which can have serious implications for biose-curity. The possibility that underwater noise from fishing vessels may promote settlement of biofouling on hulls wasinvestigated for the ascidian Ciona intestinalis. Spatial differences in biofouling appear to be correlated with spatial dif-ferences in the intensity and frequency of the noise emitted by the vessel’s generator. This correlation was confirmed inlaboratory experiments where C. intestinalis larvae showed significantly faster settlement and metamorphosis whenexposed to the underwater noise produced by the vessel generator. Larval survival rates were also significantly higher intreatments exposed to vessel generator noise. Enhanced settlement attributable to vessel generator noise may indicate thatvessels not only provide a suitable fouling substratum, but vessels running generators may be attracting larvae andenhancing their survival and growth.

Keywords: underwater sound; ascidian; larval settlement; metamorphosis; vessel noise; biofouling

Introduction

Ambient underwater sound has long been recognised as

one of the many cues used by marine organisms for ori-

entation, settlement and/or metamorphosis (Montgomery

et al. 2006; Stanley et al. 2012). Given this, it seems

equally likely that anthropogenic generated noise from

vessels in port environments may also have an influence

on the settlement of biofouling species, over and above

any natural background ambient sound. However, little

research has been conducted to verify if this is the case.

Anthropogenic noise generated by vessels typically falls

within the 6 to ~ 30,000 Hz frequency range, with the

highest intensities usually between 10 and 1,000 Hz. This

frequency range has been shown to be important for the

attraction, settlement and metamorphosis of many marine

invertebrate larvae (Montgomery et al. 2006; Götz et al.

2009; Jeffs et al. 2011). Several experimental studies

have shown that ambient underwater sound emanating

from coastal habitats can act as long-distance orientation

cues for settlement-stage crustacea (Jeffs et al. 2003,

2005; Radford et al. 2007; Stanley et al. 2010) and fish

(Tolimieri et al. 2000; Montgomery et al. 2006; Simpson

et al. 2008) attempting to locate suitable habitats. It is

also known that the planktonic larvae of many benthic

invertebrates can settle as soon as they find an appropri-

ate substratum or when conditions are favourable for col-

onisation (Glasby & Connell 1999). This facilitation may

be influenced by underwater sound.

The underwater noise produced by marine vessels

comes primarily from mechanical vibrations produced

by the engines, power transmission units and genera-

tors, from the hull interacting with the water whilst

underway and from cavitation on the rotating propeller

blades. Over the past decades ship traffic has increased,

and noise emissions account for > 90% of the acoustic

energy humans emit into the sea (Green et al. 1994).

Recent research has implicated underwater noise from

the generators of vessels as a possible cause of

increased settlement of fouling organisms through trig-

gering natural larval settlement cues (Wilkens et al.

2012). For example, settlement of larvae of the mussel,

Perna canaliculus, was significantly faster when

exposed to the underwater noise produced by a 125-m

long steel-hulled ferry (Wilkens et al. 2012). The

researchers suggested that reducing the noise emitted by

a vessel whilst at berth may help to reduce biofouling

on its hull.

Domestic fishing vessels were used as a model vector

in this study as they spend significant periods of time in

port running on generator power, and may be one of the

vectors in the spread of invasive species between ports.

Although best-practice management biofouling guidelines

are being developed for vessels in both New Zealand

and Australia, the risk of domestic spread of pest species

is still high. As vessel traffic is higher in human-

transport hubs, marine areas such as ports are regarded

*Corresponding author. Email: Serena.Wilkens@niwa.co.nz

© 2014 Taylor & Francis

Biofouling, 2014

http://dx.doi.org/10.1080/08927014.2014.919630

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

as important foci for the spread of non-indigenous

species (Floerl & Inglis 2005).

The biosecurity risks from hull fouling arise when

competent pest organisms are released into a suitable

recipient region in the form of adult life stages or

planktonic propagules and encounter a suitable settle-

ment habitat, such as a vessel hull (Floerl et al. 2004).

Due to the negative impacts of introduced marine pest

species into new non-native locations there are extensive

efforts to reduce the translocation of marine pest species

around the world (Wonham et al. 2000; Floerl et al.

2004). Vessel hull fouling may be responsible for at least

75% of the ship-mediated non-indigenous species in Port

Philip Bay, Australia (Hewitt 2004). A sampling of biota

scraped from hulls of commercial vessels in Germany

showed the presence of non-indigenous species on 96%

of the 131 ships examined (Gollasch 2002). A survey of

eight vessels in Hawaii found non-indigenous species on

the hulls of the majority of vessels, despite low levels of

overall biofouling (Godwin 2003). A single inspection of

a relatively clean passenger vessel in Australia found

several non-indigenous species in protected (niche) areas

of the ship, including the European green crab, Carcinus

maenus (Coutts et al. 2003). The fouling present on

vessels often reflects the fouling biota of the port/marina

where they spend most time, therefore movement of

fouled vessels from one location to another presents a

species translocation risk. Understanding biological and

physical variables that may influence the settlement of

hull-fouling species may be one tool in the biosecurity

management toolbox. Therefore the aim of this study

was to investigate whether spatial differences in noise

emission existed on a vessel, and what the influence

might be on the settlement of fouling species and the

overall levels of fouling present on a vessel. Evidence

that larvae respond to vessel noise was published by

Wilkens et al. (2012). This study only documented one

species (Perna canaliculus) in a controlled laboratory

environment. The present research aimed to demonstrate

that this trend is occurring in another common biofouling

species, such as ascidians (Lambert 2001; see also the

review on ascidian fouling by Aldred & Clare 2014) and

highlights spatial differences in noise intensity and

observed levels of fouling on vessels.

Materials and methods

It was not logistically possible to ‘tie up’ a number of

commercial fishing vessels to experimentally test the

effects of noise intensity on biofouling. Therefore this

study both examined in situ levels of hull fouling on

four fishing vessels (from four locations on each vessel)

and performed laboratory experiments to determine the

settlement response of a common fouling ascidian spe-

cies (Ciona intestinalis) to the different noise intensities

sampled at different locations around a vessel hull. This

approach provides a robust approach to assess the effects

of variable intensity of vessel noise on biofouling species

in a marina environment, and confirms the results by

testing in a controlled laboratory environment.

Vessel noise recording and processing

Vessel generator noises were recorded from a 25-m long

steel-hulled fishing vessel berthed in the Port of Freman-

tle, Western Australia, during February 2012. Only one

vessel was used for the recordings as the generators were

identical in the four vessels used in the fouling observa-

tions. Based on the location of the generator in this fish-

ing vessel design (Figure 1), four spatially separated hull

locations were selected for noise recordings. These were

(1) adjacent to the generator, (2) the stern, (3) the bow,

and (4) opposite the generator. At the time of recording

the vessel was operating on a ship-based generator

power supply and no other machinery was operational

during the recordings. No other vessels were operating

within the vicinity of this vessel when recordings were

taking place.

A calibrated hydrophone (HTI-96-Min, High Tech,

Inc., Long Beach, MS, USA) was used to continuously

record 5 min of underwater noise emitted by the vessel

generator at four locations around the hull (Figure 1).

The hydrophone was placed ~50 cm from the hull at

each of the four locations and lowered 2 m into the

water. During the recording phase the output was cap-

tured on a calibrated digital recorder (Edirol R09HR,

Roland Corporation, Japan).

Digital recordings were downloaded and spectral plots

were generated using the methods described in Wilkens

et al. (2012). An ANOVA was then performed to deter-

mine if there was a significant difference (p < 0.001) in

noise intensity among the four locations. The sub-samples

were then band-pass filtered into four frequency bins:

30–100, 101–500, 501–2,000, and 2,001–20,000 Hz, and

the overall mean proportion of total noise intensity was

calculated for each frequency bin. For each hull location

the proportion of total noise intensity was arcsine

Figure 1. Locations of underwater noise recordings from a25 m fishing vessel. ! denotes the location of the generator onthe vessel.

2 J.I. McDonald et al.

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

transformed and analysed using a two-way ANOVA, with

Location and Frequency Bin as factors. Significant differ-

ences between proportions of total noise intensity were

determined using the Holm–Sidak test once the ANOVA

had determined an overall significant difference among

proportions (p < 0.001).

Two-minute sequences of the recordings were

transferred onto an MP3 player (A8695, Dick Smith

Electronics, Chullora, NSW, Australia) for playback

during the ascidian larval settlement experiment.

In situ observations of level of fouling

Four 25-m fishing vessels of comparable hull design and

antifouling treatment regime were berthed together at the

time of this study. The location of the vessels, and their

generator types, were identical. The level of biofouling

present on each of the four vessel hulls was estimated

using in situ diver observations and from examination of

underwater video (Snake-Eye IIITM, Aqua Communica-

tions Inc., Waltham, MA, USA). Observations were made

at the same four locations as used for the noise recordings

(ie adjacent to the generator, the stern, the bow and oppo-

site the generator). All visual estimates of hull fouling

present on all four vessels were made by both the

co-authors McDonald and Wilkens independently using

the level of fouling scale developed by Floerl et al.

(2005). This estimate is an ordinal rank scale of the rela-

tive abundance (approximate percentage cover) of fouling

assemblage present on the hull. Both observers visually

assessed a 2 m2 area of the vertical side of each vessel at

each of the locations used for noise recordings, from the

waterline to the top of the bilge keel.

Source of ascidian larvae

C. intestinalis, like many other solitary ascidian species,

is a prolific biofouler of vessel hulls and marine infra-

structures around the world (see Fitridge et al. 2012;

Aldred & Clare 2014). Adult specimens were collected

from Lyttelton Harbour, New Zealand in January 2012

and transported to the laboratory. Individual specimens

were dissected in the laboratory. The eggs and sperm

were removed using glass Pasteur pipettes. The repro-

ductive status could be assessed visually prior to dissec-

tion to ensure only sexually mature individuals were

used (protocol adapted from Cirino et al. 2002). Differ-

ent donor specimens were used for cross fertilisation.

The eggs were placed into a Petri dish containing 25 ml

of sterile seawater and ~ 300 μl of concentrated sperm

and gently agitated to ensure mixing of gametes. The

dilution of 300 μl of sperm into 25 ml of seawater pre-

vented an excess of sperm sticking to the eggs, poten-

tially compromising insemination. One hour after

insemination, the seawater was changed to remove

surplus sperm and the Petri dishes placed at 18–20°C for

15–18 h to allow embryo development. Immediately

prior to hatching (which was confirmed using light

microscopy), embryos were randomly selected and trans-

ferred into a sterile, flat bottomed 12-well tissue culture

plate (Sarstedt, Ingle Farm, Australia). Each well con-

tained 10 ml of sterile seawater at 18°C and an individ-

ual C. intestinalis larva.

Larval settlement experiment

The larval settlement experiment consisted of five treat-

ments. The first four treatments were exposed to noise

recordings from the four different locations on the vessel:

Location 1 (adjacent to generator, starboard side), Loca-

tion 2 (opposite generator, portside), Location 3 (stern),

and Location 4 (bow). The fifth treatment was a control

(ie no vessel noise) (Figure 1). For each treatment, three

replicate water baths were used to maintain a constant

water temperature at 18°C (± 1°C) throughout the experi-

ment. Each water bath contained a single 12-well tissue

culture plate which was visually and acoustically transpar-

ent. The water baths were covered with shade cloth, pro-

viding a constant low light level, thereby reducing

interference from external light cues. Foam rubber mats

were placed under the water baths to prevent any transfer

of acoustic energy from the surrounding environment into

the experimental treatments. Prior to the commencement

of the experiment, the absence of acoustic interference in

the treatment baths was confirmed by recording from each

of the replicates within each treatment using a calibrated

hydrophone (HTI-96-MIN).

Noise in each water bath was emitted from a speaker

(SBA1500, 4Ω, 100–18,000 Hz, Koninklijke Philips

Electronics, Noord-Brabant, The Netherlands) which was

sealed within a waterproof plastic bag, placed in the bot-

tom of the water bath, and held down by a lead weight.

The speakers were connected to a MP3 player which

continuously replayed a 2 min sequence of the vessel

noise recording from the corresponding location on the

vessel. Three different 2 min sequences from each loca-

tion were used to avoid pseudo-replication by using the

same vessel recording for each replicate within the treat-

ment (Kroodsma et al. 2001). The control treatment con-

tained an underwater speaker, weighted to the bottom,

but not connected to an MP3 player. For the treatments

the experimental vessel noise replayed in the water baths

was confirmed at 140.6 dB for Location 1 (adjacent to

generator, starboard side), and at 138.8 dB for Location

2 (opposite generator, portside), 135.2 dB for Location 3

(stern), and at 127.5 dB for Location 4 (bow). Examina-

tion of the recordings of the replayed generator noise

were analysed and verified to have a similar spectral

composition to the original recording of the noise from

the vessel in port.

Biofouling 3

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

At two-hourly intervals, each tissue culture plate from

each treatment was removed from the water bath and

examined under a binocular microscope (× 40) to observe

the development of each larva. Larvae were classified

according to their progressive stages in the settlement

process, as described in Cirino et al. (2002): (1) swim-

ming; (2) immobile – larvae motionless when stimulated

by gentle suction from the tip of a 200 μm pipette, larvae

still coloured/opaque and body still intact; (3) attached –

larvae attached to the surface of the well or the meniscus

of the water by head, larvae remains attached when

gently stimulated by water movement; (4) metamorphic

stage 1 (M1) – tail at right angles to head, tail beginning

to turn transparent and starting to reabsorb, head darken-

ing/starting to turn pink, firmly attached to surface of well

or meniscus; (5) metamorphic stage 2 (M2) – tail reab-

sorption complete, pink coloration in head, larvae lobed

shaped, stalk starting to appear; or (6) dead – larvae

transparent or emaciated, head and tail starting to frag-

ment and shrink, no movement. The experiment was ter-

minated when all experimental larva had either attached

and/or metamorphosed or were dead. Data were exam-

ined to determine if there were any significant differences

between replicates within treatments. No significant dif-

ferences were found within treatments (controls or noise

treatments), therefore all replicates within a treatment

were pooled for ease of analysis and presentation of

results. Differences in larval, settlement, metamorphosis

and survival were tested using χ2 analyses.

Results

Vessel noise

The average noise intensity recorded from each location

around the vessel was measured at 140.6 dB re 1 μPa

RMS at Location 1, 138.8 at Location 2, 135.2 at Loca-

tion 3 and 127.5 at Location 4 (Figure 2). There was a

significant difference in the noise intensity among all of

the four locations (ANOVA; F = 5349.4, p < 0.001).

There was also a significant difference among locations

for frequency: the Frequency Bin factor (F = 30556.3,

p < 0.001), the Location factor (F = 25.2, p < 0.001) and

in the interaction between Frequency Bin and Location

(F = 1437.9, p < 0.001). All comparisons between

Frequency Bin and Location showed a significant differ-

ence in proportion of total noise intensity (p < 0.005)

except for the comparison between Locations 1 and 2, 2

and 3, and 1 and 4 in the highest frequency band

2,001–20,000 Hz. In general, the highest proportion of

total noise intensity occurred in the 30–100 Hz frequency

band, with the greatest of these occurring at Location 4

(Figure 3). The proportion of total noise intensity

dropped as the frequency bands increased among all

locations (Figure 3).

Level of fouling

The level of fouling (LoF) was estimated according to the

ordinal scale developed by Floerl et al. (2005). The results

indicated that biofouling was highest on all four vessels

Figure 3. Proportion of total noise intensity (Prms2) occur-ring in four frequency bands at each sampling location aroundthe vessel. Error bars represent SEs (n = 7 selections of eachrecording per location).

Figure 2. Spectral plot of vessel generator noise whenrecorded in situ from a vessel berthed in port. The y-axis isspectrum level (dB re 1 µPa/Hz) and the x-axis is frequency(Hz). The black line represents Location 1 (generator) 140.6 dBaverage between 30 and 20,000 Hz; the blue line representsLocation 2 (opposite generator) 138.8 dB average between 30and 20,000 Hz; the red line represents Location 3 (stern)135.2 dB average between 30 and 20,000 Hz; the green linerepresents Location 4 (bow) 127.5 dB average between 30 and20,000 Hz.

4 J.I. McDonald et al.

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

examined, at the location closest to the generator

(Figure 4). This site was also confirmed to have the high-

est intensity of noise when analysed. The biofouling level

(relative abundance of fouling present on the vessel sur-

face) decreased with increasing distance away from the

noise source (ie the generator), with the bow showing the

lowest level of biofouling. All four vessels examined

showed a similar trend, with the highest overall LoF at

the generator and lowest at the bow. Intermediate LoF

ranks were determined for the sites opposite the generator

and at the stern. Both observers consistently scored the

same fouling estimate for every vessel and location sam-

pled. Biofouling consisted predominantly of colonial and

solitary ascidians (Polycitor sp., Sigillina sp., Botrylloides

sp., Styela sp.) bryozoans (Bugula sp., Zoobotryon

verticillatum, Watersipora subtorquata, W. arcuata), ser-

pulid polychaetes (unidentified) and porifera (unidenti-

fied). Although specific species data are not presented

here, a total of 24 morphotypes were identified from the

biofouling samples, of which four were confirmed to be

non-indigenous species and three were cryptogenic.

Larval settlement experiment

At the commencement of the experiment, all C. intestinalis

larvae were swimming and viable when introduced into the

experimental chambers.

Settlement

In the first 2–4 h of the experiment larvae in the noise

treatments began to settle. These larvae then continued

to settle at a significantly faster rate than the larvae in

the control treatment (χ2: 71.42; χ2 critical value: 9.48).

By 6 h, half of the larvae in the noise treatments had set-

tled and all surviving larvae had settled by 18–20 h

(Figure 5). This contrasts markedly with the controls,

where larvae did not commence settlement until 6 h into

the experiment and showed a much slower, more

staggered settlement pattern (Figure 5). It took ~ 15 h for

50% of the larvae in the controls to settle and a total of

26 h for all the remaining surviving larvae to settle

(Figure 5).

Metamorphosis and survival

While initial settlement commenced as early as 2 h meta-

morphosis to M2 did not occur until 6–8 h into the experi-

ment. All larvae (in both the controls and treatments)

began to undergo metamorphosis at approximately the

same time. However the rate at which larvae continued to

develop through to metamorphic stage 2 (M2) was signifi-

cantly different between the noise treatments and silent

control (χ2: 20.75; χ2 critical value: 9.48). In all the noise

treatments, the larvae displayed an exponential rate of

metamorphosis to stage M2 between 10 and 20 h after the

commencement of the trial, and 90–100% of the total

larvae developed through to this stage. In the control only

66% larvae survived and succeeded to develop to M2,

and it took ~ 22 h for all surviving larvae to undergo this

metamorphosis. The rate of metamorphosis was also more

variable in the control, with groups of larvae settling

intermittently over time.

Vessel noise had a significant influence upon the sur-

vival of C. intestinalis larvae. Larvae subjected to all

5

4

3

2

1

0

Lev

el o

f fo

uli

ng e

stim

ate

Location 1 (gen) Location 2 (opp gen) Location 3 (stern) Location 4 (bow)

Figure 4. Level of fouling rank applied to each to each loca-tion on the four vessels inspected. Location 1 (generator) wasrecorded immediately adjacent to the generator, Location 2(opposite generator) was directly opposite generator, Location 3(stern) at stern of vessel, and Location 4 (bow) at bow ofvessel. Note overlap between vessels is due to same foulingestimates at each location. (Fouling rank estimated as per Floerlet al. 2005.). Vessel 1, Vessel 2, Vessel 3, and Vessel 4

Pro

port

ion o

f su

rviv

ing l

arvae

set

tled

100

90

80

70

60

50

40

30

20

10

00 2 4 6 8 10 12 14

Time (h)

Control

Location 1 (gen)

Location 2 (opp gen)

Location 3 (stern)

Location 4 (bow)

16 18 20 22 24 26

Figure 5. Proportion (%) of surviving ascidian larvae settledover time (h) for each treatment. Location 1 (generator) wasrecorded immediately adjacent to the generator; Location 2(opposite generator) was directly opposite generator; Location3 (stern) at the stern of the vessel; and Location 4 (bow) at thebow of the vessel. No significant differences were found withintreatments (controls or sound treatments), so all replicateswithin a treatment were pooled for ease of analysis and presen-tation of results.

Biofouling 5

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

vessel noise treatments had significantly greater survival

rates than those in the silent treatment (χ2: 56.55; χ2 criti-

cal value: 9.48). In noise treatment 1 (immediately adja-

cent to the generator) and 2 (directly opposite the

generator), survival was 100%. In noise treatments 3 (at

the stern) and 4 (at the bow), survival was 94 and 88%

respectively. However in the control (which was silent),

survival was only 66%.

Discussion

The research presented here provides evidence that

spatial differences in noise emission exist at different

locations on a vessel, and demonstrates a relationship

between these noise emissions and in situ biofouling.

The study experimentally demonstrated that when

exposed to increased levels of vessel noise, larval settle-

ment, metamorphosis and survival increases, compared

with a silent control.

Spatial distribution of sound on the vessel

The intensity of the vessel generator noise was con-

firmed to be highest at the site closest to the generator

(Location 1) and decreased away from the noise source

(with Location 4 with the lowest intensity). The propor-

tion of total noise intensity in each frequency bin dif-

fered significantly around the vessel. However, at all

sites on the vessel low frequency (30–100 Hz frequency

bin) dominated the noise spectra. Therefore it appears

that the low frequency noise may be the important factor

which is driving increased larval settlement and survival.

One possible explanation for increased settlement on

vessel hulls is that low frequency vessel noise may

reduce the boundary layer near the surface of the hull

and therefore make the settlement surface more accessi-

ble to larvae. However more research is required to

ascertain whether this is a behavioural or physical mech-

anism which is influencing larval settlement. Noise is

certainly not the only larval settlement or metamorphosis

cue and a major function of the settlement stage is in

response to a range of environmental inputs (Rittschof

et al. 1998). Chemical cues (Alfaro et al. 2006), surface

rugosity (Alfaro et al. 2004), light and temperature

(Bayne 1964; Carl et al. 2011) are also important. How-

ever, noise is likely to be an important factor in ports

and marinas due to its ability to traverse relatively large

distances and through turbid, turbulent waters which

may restrict or alter chemical or visual cues.

Level of fouling

The influence of the spatial distribution of sound inten-

sity was reflected in the observed LoF ranks at each of

the four locations on the vessel. Observed levels of

biofouling (applied according to the LoF rank; Floerl

et al. 2005) were higher on the fishing vessel in the sites

closest to the generator. The observed LoF data coupled

with the spatial distribution of sound around the vessel

suggest that low frequency noise is influencing a range

of biofouling species and the trend is more widespread

than initially thought.

Laboratory experiments

Results from this research clearly demonstrated that set-

tlement, metamorphosis and survival in C. intestinalis

larvae are influenced by vessel generator noise in labora-

tory experiments. C. intestinalis larvae exposed to vessel

generator noise settled and metamorphosed significantly

faster than larvae which were not exposed to any genera-

tor noise. Approximately 50% of the surviving larvae

exposed to vessel generator noise had settled 6 h after

the commencement of the experiment, with the remain-

ing larvae all settling by 18 h. This contrasts markedly

with the control where it took 15 h for 50% of surviving

larvae to settle and a total of 26 h for all surviving larvae

to settle. The enhanced settlement attributable to vessel

generator noise may indicate that stationary vessels not

only provide a suitable substratum for the settlement of

fouling species (Godwin 2003) but the noise from vessel

generators may be attracting larvae and enhancing their

survival. Substratum vibration resulting from vessel

sound may also contribute to increased settlement in that

the physical vibration influences the behaviour of the

larvae. Ascidian larvae attach to surfaces when there is

contact with the apical papillae so it is possible that

vibration of the settlement surface is sufficient to stimu-

late adhesive release (Rittschof et al. 1998; Aldred &

Clare 2014) (a physical rather than behavioural mecha-

nism, driven by low frequency sound vibration). Many

larvae have mechanosensory ability and can detect vibra-

tion, possibly signalling permanence to the larvae, that

is, the likelihood that a surface will last sufficiently long

for growth, maturation and reproduction (Rittschof et al.

1998). Clearly, further investigation into the mechanism

by which vessel sound influences larvae settlement

behaviour is required.

In its native range, natural populations of C. intestinalis

are found attached to rocky substrata (Petersen & Riisgard

1992), suggesting that ambient underwater sound produced

by rocky coastal habitats may be one important cue for the

settlement of C. intestinalis larvae. C. intestinalis also has

a tendency to attach to the hulls of ships and foul pylons

and other marine infrastructure (McDonald 2004) of noisy

ports and marinas around the globe. Although

C. intestinalis was not present on the vessels sampled, this

is an excellent candidate for laboratory trials as it is such a

prolific biofouler. In this experiment, not only did noise

have a significant influence on settlement, but once settled,

6 J.I. McDonald et al.

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

larvae underwent metamorphosis at a much faster rate

when exposed to vessel generator noise. Development to

M2 stage (recognised as when a larvae has settled,

reabsorbed the tail and developed into an adult form) was

achieved in 60% of the larvae exposed to the noise

treatments (over a 12 h period), compared with only 20%

in the control treatment over the same period. Larvae

subjected to the two highest intensity noise treatments

(immediately adjacent to the generator and opposite the

generator) had a 100% survival rate compared to a maxi-

mum survival of only 66% in the silent control. The

enhanced survival (89–100% across all noise treatments)

and development of larvae when influenced by vessel

noise represents a significant risk in terms of vessel medi-

ated spread of invasive species. While sound reception has

been reported in some cephalopod molluscs (Packard et al.

1990; Williamson 1995; Mooney et al. 2010), to the

authors’ knowledge, the discovery that ascidian larvae can

detect and respond to vessel noise is the first description of

an auditory response in ascidians, which is a major group

in terms of biomass in shallow aquatic communities and

also biofouling assemblages (Lambert & Lambert 2003).

This response to vessel noise has been previously

documented in mussel larvae (Wilkens et al. 2012),

suggesting the response is active in other fouling taxa, but

until now this theory has not been documented in fouling

species such as ascidians.

As the density of the water transmits acoustic energy

very efficiently and over greater distances than in air, the

effects of underwater noise in a port environment may

extend over vast distances. The transmission of vessel

generator noise may also be increasing the area over

which larvae are attracted to a vessel. The vessel noises

recorded in this study had a noise intensity of between

127.5 dB and 140.6 dB re 1 μPa RMS (and between

50–24,000 Hz) which is within the range of sound inten-

sities recorded for natural reefs (Radford et al. 2008a,

2008b, 2010; Simpson et al. 2011). Assuming cylindrical

spreading of noise from the ship hull, which is appropri-

ate in the shallow water where vessels are berthed, then

the noise would occur at ~ 500 m out from the hull. Thus

if a clean vessel enters an infected port and runs on gen-

erator power it could be attracting fouling pest species

from a ~ 500 m radius. It has been demonstrated that

vessels acquire the majority of hull fouling while moored

in coastal ports, primarily because these environments

act as hubs for domestic and international shipping

movements (Carlton 1987). In 2002, Callow and Callow

reported that biofouling was a large and continuing prob-

lem, one that costs the US Navy up to $1 billion per

annum to manage. Shultz et al. (2011) also document

massive economic costs and highlight the fact that for a

single vessel class within the US Navy fleet biofouling

costs US$56 million.

Interestingly, there was no significant difference in

larval settlement, metamorphosis or survival rates between

the different noise treatments. This contrasts with the

observed level of fouling present on the vessels, where

significantly greater fouling cover was associated with

higher noise intensity (immediately adjacent to the genera-

tor and decreasing with distance from the noise source). A

possible explanation for this may be that C. intestinalis

larvae are not able to discriminate between the different

intensities associated with the four noise treatments. Many

other species such as crabs, fish and coral have been

documented to respond to ambient noise within the

100–1,000 Hz range (Montgomery et al. 2006; Radford

et al. 2007; Vermeij et al. 2010), which is certainly within

the range tested in these settlement experiments, and pos-

sibly other species are more capable of discriminating

between noise intensity, thereby explaining the observed

differences in biofouling on the vessel hulls.

Conclusion

This research suggests that vessels running generators in

ports may be promoting the settlement and survival of

marine pest species on their submerged surfaces. In the

absence of light, temperature or chemical cues from con-

specifics, the rates of settlement, metamorphosis and sur-

vival are significantly increased in C. intestinalis larvae

when exposed to vessel noise. It is, of course, more cost-

effective to prevent invasive species from establishing

themselves than it is to try to eradicate them once they

have become established. The goal therefore should be to

minimise the arrival of new species. As ports are apparent

havens for introduced marine species, vessels running

generators in ports seem to be at higher risk of attracting

biofouling based on the results presented here. Mecha-

nisms to ameliorate this biofouling potential should be

further investigated and tested. The dampening of genera-

tor noise may reduce the rate at which fouling accrues, or

vessels may be able to run on land-based power supplies

as opposed to ship-based generators. Alternatively there

may be a frequency at which vessel noise either ceases to

be a concern or can be ‘tuned’ to repel fouling. These

areas need to be further investigated.

Acknowledgements

The authors thank Kailis Marine (Terry Hewitt) for access to vesselsin Fremantle, Caroline Williams (NIWA) for collecting Cionaintestinalis adults, the University of Auckland Leigh MarineLaboratory for provision of laboratory space and equipment,Samantha Bridgwood for assistance with fieldwork and internalreviewers for constructive feedback and review of the manuscript.This research was funded by NIWA under the Innovative Seed Fund(2011/2012) and Coasts and Oceans Programme 4 – MarineBiosecurity (2014/15 SCI), the Glenn Family Foundation andWestern Australian Department of Fisheries. The authors wish tothank Wayne Young and Dan Pederson (Dampier Port Authority)for their support and enthusiasm.

Biofouling 7

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

References

Aldred N, Clare AS. 2014. Mini-review: impact and dynamicsof surface fouling by solitary and compound ascidians.Biofouling. 30:259–270.

Alfaro AC, Jeffs AG, Creese RG. 2004. Bottom-drifting algal/mussel spat association along sandy coastal region innorthern New Zealand. Aquaculture. 241:269–290.

Alfaro AC, Copp BR, Appleton DR, Kelly S, Jeffs AG. 2006.Chemical cues promote settlement in larvae of the green-lipped mussel, Perna canaliculus. Aquacult Int. 14:405–412.

Bayne BL. 1964. The responses of the larvae of Mytilus edulisL. to light and to gravity. Oikos. 15:162–174.

Callow ME, Callow JA. 2002. Marine biofouling: a stickyproblem. Biologist. 49:1–5.

Carl C, Poole AJ, Vucko MJ, Williams MR, Whalan S, de Nys R.2011. Optimising settlement assays of pediveligers andplantigrades of Mytilus galloprovincialis. Biofouling. 27:859–868.

Carlton JT. 1987. Patterns of transoceanic marine biologicalinvasions in the Pacific Ocean. Bull Mar Sci. 41:452–465.

Cirino P, Toscano A, Caramiello D, Macina A, Miraglia V,Monte A 2002. Laboratory culture of the ascidian Cionaintestinalis (L.): a model system for molecular developmen-tal biology research. Mar Mod Elec Rec. [cited 2012 Oct10]. Available from: http://www.mbl.edu/html/BB/MMER/CIR/CirTit.html

Coutts ADM, Moore KM, Hewitt CL. 2003. Ships’ sea-chests:an overlooked transfer mechanism for non-indigenous mar-ine species? Mar Poll Bull. 46:1504–1515.

Fitridge I, Dempster T, Guenther J, DeNys R. 2012. Theimpact and control of biofouling in marine aquaculture: areview. Biofouling. 28:649–669.

Floerl O, Pool TK, Inglis GJ. 2004. Positive interactionsbetween nonindigenous species facilitate transport byhuman vectors. Ecol Appl. 14:1724–1736.

Floerl O, Inglis GJ, Hayden BJ. 2005. A risk-based predictivetool to prevent accidental introductions of non-indigenousmarine species. Environ Manage. 35:765–778.

Glasby TM, Connell SD. 1999. Urban structures as marinehabitats. Ambio. 28:595–598.

Godwin LS. 2003. Hull fouling of maritime vessels as a path-way for marine species invasions to Hawaiian islands. Bio-fouling. 19:123–131.

Gollasch S. 2002. The importance of ship hull fouling as a vec-tor of species introductions into the North Sea. Biofouling.18:105–121.

Götz T, Hastie G, Hatch LT, Raustein O, Southall BL, Tasker M,Thomsen F. 2009. Overview of the impacts of anthropogenicunderwater sound in the marine environment. London:OSPAR Commission; p. 134.

Green DM, DeFerrari HA, McFadden D, Pearse JS, PopperAN, Richardson WJ, Ridgway SH, Tyack PL. 1994.Low-frequency sound and marine mammals: currentknowledge and research needs. Washington (DC): NationalAcademy Press NRC Report.

Hewitt CL. 2004. Invasive species – vectors and managementstrategies. Book Review. J Exp Mar Biol Ecol. 311:186–187.

Jeffs AG, Radford C, Stanley JA. 2011. Ecological effects ofacoustic pollution in the marine environment. In: Allodi S,Nazari E, editors. Pollution in the marine environment.New Delhi: Research Signpost Publishing; p. 176–198.

Jeffs A, Tolimieri N, Montgomery JC. 2003. Crabs on cue forthe coast: the use of underwater sound for orientation bypelagic crab stages. Mar Freshwater Res. 54:841–845.

Jeffs AG, Montgomery JC, Tindle CT. 2005. How do spinylobster post-larvae find the coast? New Zeal J Mar Fresh.39:605–617.

Kroodsma DE, Byers BE, Goodale E, Johnson S, Liu WC.2001. Pseudoreplication in playback experiments, revisiteda decade later. Anim Behav. 61:1029–1033.

Lambert G. 2001. A global review of ascidian introduction andtheir possible impact on the endemic fauna. In: Sawada R,Yokosawa H, Lambert CC, editors. The biology of ascidians.Heidelberg, New York, Tokyo: Springer-Verlag; p.249–257.

Lambert CG, Lambert G. 2003. Persistence and differentialdistribution of nonindigenous ascidians in harbours of theSouthern California Bight. Mar Ecol Prog Ser. 259:145–161.

McDonald JI. 2004. The invasive pest species Ciona intestinalis(Linnaeus, 1767) reported in a harbour in southern WesternAustralia. Mar Poll Bull. 49:854–874.

Montgomery JC, Jeffs AG, Simpson S, Meekan M, Tindle C.2006. Sound as an orientation cue for the pelagic larvae ofreef fish and decapod crustaceans. Adv Mar Biol. 51:143–196.

Mooney TA, Hanlon RT, Christensen-Dalsgaard J, Madsen PT,Ketten DR, Nachtigall PE. 2010. Sounds detection by thelongfin squid (Loligo pealeii) studied with auditory evokedpotentials: sensitivity to low-frequency particle motion andnot pressure. J Exp Biol. 213:3748–3759.

Packard A, Karlsen HE, Sand O. 1990. Low frequency hearingin cephalopods. J Comp Physiol. 166:501–505.

Peterson JK, Riisgard HU. 1992. Filtration capacity of theascidian Ciona intestinalis and its grazing impact in ashallow fjord. Mar Ecol Prog Ser. 88:9–17.

Radford CA, Jeffs AG, Montgomery JC. 2007. Directionalswimming response of five species of crab postlarvae inresponse to reef sound. Bull Mar Sci. 80:69–378.

Radford CA, Jeffs AG, Tindle CT, Montgomery JC. 2008a.Resonating sea urchin skeletons create coastal choruses.Mar Ecol Prog Ser. 362:37–43.

Radford CA, Jeffs AG, Tindle CT, Montgomery JC. 2008b.Temporal patterns in ambient noise of biological originfrom a shallow water temperate reef. Oecologia. 156:921–929.

Radford CA, Stanley JA, Tindle CT, Montgomery JC, JeffsAG. 2010. Localised coastal habitats have distinct under-water sound signatures. Mar Ecol Prog Ser. 401:21–29.

Rittschof D, Forward RB, Cannon G, Welch JM, McClary Jr. M,Holm ER, Clare AS, Conova S, McKelvey LM, Bryan P,Van Dover CL. 1998. Cues and context: larval responses tophysical and chemical cues. Biofouling. 12:31–44.

Schultz MP, Bendick JA, Holm ER, Hertel WM. 2011. Eco-nomic impact of biofouling on a naval surface ship. Bio-fouling. 27:87–98.

Simpson SD, Meekan MG, Jeffs AG, Montgomery JC,McCauley RD. 2008. Settlement-stage coral reef fishes pre-fer the higher frequency invertebrate-generated audiblecomponent of reef noise. Anim Behav. 75:1861–1868.

Simpson SD, Radford AN, Tickle EJ, Meekan MG, Jeffs AG.2011. Adaptive avoidance of reef noise. PLoS ONE. 6:e16625.

Stanley J, Radford C, Jeffs AG. 2010. Induction of settlementin crab megalopae by ambient underwater reef sound.Behav Ecol. 21:113–120.

Stanley J, Radford C, Jeffs AG. 2012. Location, location, loca-tion: finding a suitable home among the noise. Proc R SocBiol Sci Ser B. 279:3622–3631.

8 J.I. McDonald et al.

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu

Tolimieri N, Jeffs AG, Montgomery JC. 2000. Ambient soundas a cue for navigation by the pelagic larvae of reef fishes.Mar Ecol Prog Ser. 207:219–224.

Vermeij MJA, Marhaver KL, Huijbera CM, Nagelkerken I,Simpson SD. 2010. Coral larvae move toward reef sounds.PLoS ONE. 5:e10660.

Wilkens SL, Stanley JA, Jeffs AG. 2012. Induction of settle-ment in mussel (Perna canaliculus) larvae by vessel noise.Biofouling. 28:65–72.

Williamson R. 1995. A sensory basis for orientation incephalopods. J Mar Biol Assoc UK. 75:83–92.

Wonham MJ, Carlton JT, Ruiz GM, Smith LD. 2000. Fish andships: relating dispersal frequency to success in biologicalinvasions. Mar Biol. 136:1111–1121.

Biofouling 9

N OPQR OST UTV WXN UYSZ[ \UQ[]̂_̀ ab UZ` Ua cde_fghi cXj ka[` Q\lT OQSRTi S[mn opqrst kQUsrmu