Journal of Fish Biology (2015) doi:10.1111/jfb.12651, available online at wileyonlinelibrary.com
The influence of various reef sounds on coral-fish larvae behaviour E. Parmentier*†, L. Berten*‡, P. Rigo*, F. Aubrun‡, S. Nedelec‡§, S. D. Simpson§‖ and D. Lecchini‡¶ *Laboratoire de Morphologie Fonctionnelle et Evolutive, AFFISH-RC, Institut de chimie, Bât. B6c, Université de Liège, B-4000 Liège, Belgium, ‡USR 3278 CNRS-EPHE, CRIOBE, 97829 Moorea, French Polynesia, §School of Biological Sciences & Cabot Institute, University of Bristol, Woodland Road, Bristol, BS8 1UG, U.K., ‖Biosciences, College of Life & Environmental Sciences, University of Exeter, EX4 4QD, U.K. and ¶Laboratoire d’Excellence "CORAIL", 97829 Moorea, French Polynesia (Received 7 October 2014, Accepted 28 January 2015) The swimming behaviour of coral-reef fish larvae from 20 species of 10 different families was tested under natural and artificial sound conditions. Underwater sounds from reef habitats (barrier reef, fringing reef and mangrove) as well as a white noise were broadcasted in a choice chamber experiment. Sixteen of the 20 species tested significantly reacted to at least one of the habitat playback conditions, and a range of responses was observed: fishes were (1) attracted by a single sound but repelled by none (e.g. white-banded triggerfish Rhinecanthus aculeatus was attracted by the barrier-reef sound), (2) repelled by one or more sounds but attracted by none (e.g. bridled cardinalfish Pristiapogon fraenatus was repelled by the mangrove and the bay sounds), (3) attracted by all sounds (e.g. striated surgeonfish Ctenochaetus striatus), (4) attracted and repelled by several sounds (e.g. whitetail dascyllus Dascyllus aruanus was attracted by the barrier-reef sound and repelled by the mangrove sound) and (5) not influenced by any sound (e.g. convict surgeonfish Acanthurus triostegus). Overall, these results highlight two settlement strategies: a direct selection of habitats using sound (45% of the species), or a by-default selection by avoidance of certain sound habitats (35%). These results also clearly demonstrated the need to analyse the influence of sounds at the species-specific level since congeneric and confamilial species can express different behaviours when exposed to the same sounds. © 2015 The Fisheries Society of the British Isles
Key words: acoustic cues; coral reef; habitat; orientation; settlement; teleosts.
INTRODUCTION Coral-reef fishes have a complex life cycle, characterized by a pelagic larval stage, followed by benthic juvenile and adult stages (Leis & McCormick, 2002). Reef colonization takes place mainly during the night and corresponds to the passage of the larvae from the ocean to a reef (Dufour & Galzin, 1993), quickly followed by the settlement in an adequate habitat (Lecchini, 2005). These steps are generally related to morphological and physiological modifications to adapt to important environmental changes related to the transition from pelagic life in open sea to benthic life on the reef †Author to whom correspondence should be addressed. Tel.: +3243665024; email: [email protected]
1 © 2015 The Fisheries Society of the British Isles
2
E . PA R M E N T I E R E T A L.
(McCormick et al., 2002; Parmentier et al., 2004). During the first few hours following the colonization and settlement, the mortality rate is high, with up to 60% mortality on the first day post-colonization (Doherty et al., 2004), implying that the ability to quickly find and settle in an appropriate habitat is highly important for larval survival (Kaufman et al., 1992; Lecchini, 2005). The settlement habitat is known to be chosen according to various characteristics, such as conspecific presence, quality of habitat or absence of predators (Öhman et al., 1998; Lecchini et al., 2007a; Ben-Tzvi et al., 2009). Orientation towards suitable habitats appears possible with the use of chemical (Atema et al., 2002; Lecchini et al., 2005; Arvedlund & Takemura, 2006; Dixson et al., 2008), visual (Leis & Carson-Ewart, 2003; Lecchini et al., 2007b; Igulu et al., 2011) and acoustic cues (Tolimieri et al., 2000; Leis et al., 2003; Simpson et al., 2004; Holles et al., 2013). The influence of reef sound on settling fishes has become a subject of particular interest in the past few years, since Stobutzki & Bellwood (1998) showed that fish larvae could use sound as a cue for orientation. Several experiments using choice chambers have been used in recent years and highlighted an attractive effect of broadcasted reef noise (Tolimieri et al., 2004) and the influence of artificial tone noise on fish behaviour (Simpson et al., 2010; Holles et al., 2013). The use of sound to locate habitat has also been recently shown at Lizard Island (Australia) where juveniles are significantly more attracted to patch reefs with lagoon or fringing-reef sound broadcast than to patch reefs with no playback sound (Radford et al., 2011). In all these studies, differences were observed at the family level. For example, Pomacentridae were preferentially found on patch reefs broadcasting fringing-reef sound than on other treatments (Radford et al., 2011). The study of responses at the family level, however, is potentially oversimplified. The number of species studied in the same taxa can considerably modify the results when they are all pooled to consider the family level. In the above studies, Pomacentridae were, overall, significantly attracted by a particular sound playback, but it does not imply that all Pomacentridae species show the same kind of behaviour because species have different ecological niches, meaning that they may not be sensitive in the same way to different cues. This remark is particularly important as the sounds emitted from different locations (fringing reef, back reef and lagoon) have revealed differences in the temporal and frequency composition and in sound levels (Radford et al., 2014). At this stage, species-specific reactions to different kinds of reef sounds are not known. In this study, the influence of broadcasted natural sounds (fringing-reef, barrier-reef and mangrove sounds) and artificial white noise on the fish behaviour during their settlement stage was tested. In particular, the ability of 21 fish species from 10 families to use the sounds of three specific habitats as a cue for the selection of their settlement habitat is determined. Moreover, it is shown that the study at the family level can mask behaviour at the species-specific level.
MATERIALS AND METHODS This study was conducted at Moorea (French Polynesia; 17∘ 31′ S; 149∘ 51′ W) from February to April 2011 and 2012 using fish larvae captured in crest nets (Lecchini et al., 2006). A total of 1449 fish larvae belonging to 21 species and 10 families were tested (see Table I), with a sample size of six to 25 fishes per species for each sound treatment. Due to a lack of larvae, one test (the white-noise test) was not conducted on whitetail dascyllus Dascyllus aruanus (L. 1758) and Bennet’s sharpnose buffer Canthigaster bennetti (Bleeker 1854). The total lengths
© 2015 The Fisheries Society of the British Isles, Journal of Fish Biology 2015, doi:10.1111/jfb.12651
© 2015 The Fisheries Society of the British Isles, Journal of Fish Biology 2015, doi:10.1111/jfb.12651
Acanthurus triostegus Ctenochaetus striatus Apogon doryssa Apogon sp. Ostorhinchus angustatus Pristiapogon exostigma Pristiapogon fraenatus Rhinecanthus aculeatus Chaetodon citrinellus Valenciennea strigata Myripristis kuntee Neoniphon argenteus Sargocentron microstoma Sargocentron spiniferum Ptereleotris microlepsis Abudefduf sexfasciatus Chromis viridis Chrysiptera leucopoma Dascyllus aruanus Scorpaenodes guamensis Canthigaster bennetti
None All – BR (
The influence of various reef sounds on coral-fish larvae behaviour E. Parmentier*†, L. Berten*‡, P. Rigo*, F. Aubrun‡, S. Nedelec‡§, S. D. Simpson§‖ and D. Lecchini‡¶ *Laboratoire de Morphologie Fonctionnelle et Evolutive, AFFISH-RC, Institut de chimie, Bât. B6c, Université de Liège, B-4000 Liège, Belgium, ‡USR 3278 CNRS-EPHE, CRIOBE, 97829 Moorea, French Polynesia, §School of Biological Sciences & Cabot Institute, University of Bristol, Woodland Road, Bristol, BS8 1UG, U.K., ‖Biosciences, College of Life & Environmental Sciences, University of Exeter, EX4 4QD, U.K. and ¶Laboratoire d’Excellence "CORAIL", 97829 Moorea, French Polynesia (Received 7 October 2014, Accepted 28 January 2015) The swimming behaviour of coral-reef fish larvae from 20 species of 10 different families was tested under natural and artificial sound conditions. Underwater sounds from reef habitats (barrier reef, fringing reef and mangrove) as well as a white noise were broadcasted in a choice chamber experiment. Sixteen of the 20 species tested significantly reacted to at least one of the habitat playback conditions, and a range of responses was observed: fishes were (1) attracted by a single sound but repelled by none (e.g. white-banded triggerfish Rhinecanthus aculeatus was attracted by the barrier-reef sound), (2) repelled by one or more sounds but attracted by none (e.g. bridled cardinalfish Pristiapogon fraenatus was repelled by the mangrove and the bay sounds), (3) attracted by all sounds (e.g. striated surgeonfish Ctenochaetus striatus), (4) attracted and repelled by several sounds (e.g. whitetail dascyllus Dascyllus aruanus was attracted by the barrier-reef sound and repelled by the mangrove sound) and (5) not influenced by any sound (e.g. convict surgeonfish Acanthurus triostegus). Overall, these results highlight two settlement strategies: a direct selection of habitats using sound (45% of the species), or a by-default selection by avoidance of certain sound habitats (35%). These results also clearly demonstrated the need to analyse the influence of sounds at the species-specific level since congeneric and confamilial species can express different behaviours when exposed to the same sounds. © 2015 The Fisheries Society of the British Isles
Key words: acoustic cues; coral reef; habitat; orientation; settlement; teleosts.
INTRODUCTION Coral-reef fishes have a complex life cycle, characterized by a pelagic larval stage, followed by benthic juvenile and adult stages (Leis & McCormick, 2002). Reef colonization takes place mainly during the night and corresponds to the passage of the larvae from the ocean to a reef (Dufour & Galzin, 1993), quickly followed by the settlement in an adequate habitat (Lecchini, 2005). These steps are generally related to morphological and physiological modifications to adapt to important environmental changes related to the transition from pelagic life in open sea to benthic life on the reef †Author to whom correspondence should be addressed. Tel.: +3243665024; email: [email protected]
1 © 2015 The Fisheries Society of the British Isles
2
E . PA R M E N T I E R E T A L.
(McCormick et al., 2002; Parmentier et al., 2004). During the first few hours following the colonization and settlement, the mortality rate is high, with up to 60% mortality on the first day post-colonization (Doherty et al., 2004), implying that the ability to quickly find and settle in an appropriate habitat is highly important for larval survival (Kaufman et al., 1992; Lecchini, 2005). The settlement habitat is known to be chosen according to various characteristics, such as conspecific presence, quality of habitat or absence of predators (Öhman et al., 1998; Lecchini et al., 2007a; Ben-Tzvi et al., 2009). Orientation towards suitable habitats appears possible with the use of chemical (Atema et al., 2002; Lecchini et al., 2005; Arvedlund & Takemura, 2006; Dixson et al., 2008), visual (Leis & Carson-Ewart, 2003; Lecchini et al., 2007b; Igulu et al., 2011) and acoustic cues (Tolimieri et al., 2000; Leis et al., 2003; Simpson et al., 2004; Holles et al., 2013). The influence of reef sound on settling fishes has become a subject of particular interest in the past few years, since Stobutzki & Bellwood (1998) showed that fish larvae could use sound as a cue for orientation. Several experiments using choice chambers have been used in recent years and highlighted an attractive effect of broadcasted reef noise (Tolimieri et al., 2004) and the influence of artificial tone noise on fish behaviour (Simpson et al., 2010; Holles et al., 2013). The use of sound to locate habitat has also been recently shown at Lizard Island (Australia) where juveniles are significantly more attracted to patch reefs with lagoon or fringing-reef sound broadcast than to patch reefs with no playback sound (Radford et al., 2011). In all these studies, differences were observed at the family level. For example, Pomacentridae were preferentially found on patch reefs broadcasting fringing-reef sound than on other treatments (Radford et al., 2011). The study of responses at the family level, however, is potentially oversimplified. The number of species studied in the same taxa can considerably modify the results when they are all pooled to consider the family level. In the above studies, Pomacentridae were, overall, significantly attracted by a particular sound playback, but it does not imply that all Pomacentridae species show the same kind of behaviour because species have different ecological niches, meaning that they may not be sensitive in the same way to different cues. This remark is particularly important as the sounds emitted from different locations (fringing reef, back reef and lagoon) have revealed differences in the temporal and frequency composition and in sound levels (Radford et al., 2014). At this stage, species-specific reactions to different kinds of reef sounds are not known. In this study, the influence of broadcasted natural sounds (fringing-reef, barrier-reef and mangrove sounds) and artificial white noise on the fish behaviour during their settlement stage was tested. In particular, the ability of 21 fish species from 10 families to use the sounds of three specific habitats as a cue for the selection of their settlement habitat is determined. Moreover, it is shown that the study at the family level can mask behaviour at the species-specific level.
MATERIALS AND METHODS This study was conducted at Moorea (French Polynesia; 17∘ 31′ S; 149∘ 51′ W) from February to April 2011 and 2012 using fish larvae captured in crest nets (Lecchini et al., 2006). A total of 1449 fish larvae belonging to 21 species and 10 families were tested (see Table I), with a sample size of six to 25 fishes per species for each sound treatment. Due to a lack of larvae, one test (the white-noise test) was not conducted on whitetail dascyllus Dascyllus aruanus (L. 1758) and Bennet’s sharpnose buffer Canthigaster bennetti (Bleeker 1854). The total lengths
© 2015 The Fisheries Society of the British Isles, Journal of Fish Biology 2015, doi:10.1111/jfb.12651
© 2015 The Fisheries Society of the British Isles, Journal of Fish Biology 2015, doi:10.1111/jfb.12651
Acanthurus triostegus Ctenochaetus striatus Apogon doryssa Apogon sp. Ostorhinchus angustatus Pristiapogon exostigma Pristiapogon fraenatus Rhinecanthus aculeatus Chaetodon citrinellus Valenciennea strigata Myripristis kuntee Neoniphon argenteus Sargocentron microstoma Sargocentron spiniferum Ptereleotris microlepsis Abudefduf sexfasciatus Chromis viridis Chrysiptera leucopoma Dascyllus aruanus Scorpaenodes guamensis Canthigaster bennetti
None All – BR (
