Homeward Sound

See allHide authors and affiliations

Science  08 Apr 2005:
Vol. 308, Issue 5719, pp. 221
DOI: 10.1126/science.1107406


Despite spending weeks at sea as larvae, potentially scattered over many kilometers, young coral reef fish find suitable settlement habitat and in some cases return to their natal reefs. We report that some dominant families of larval reef fish use the sounds made by fish and shrimp resident on reefs to help them locate and settle on reefs and that some fish groups use specific components of the reef sound to guide their behavior. These findings could offer potential for active management of reef fisheries.

Most reef populations are replenished with recruits that settle out from an initially pelagic existence. The larvae of nearly all coral reef fish develop at sea for weeks to months before settling back to reefs as juveniles. Although larvae have the potential to disperse great distances, recent studies show a substantial portion recruit back to their natal reefs (1, 2). Larvae are not passively dispersed but develop a high level of swimming competence (3). How they use these capabilities to influence their dispersal is an open question. We show here that recruits respond actively to reef sounds, potentially providing a valuable management tool for the future.

Since the discovery that reef fish larvae are accomplished swimmers, focus has shifted to identifying cues that may influence their orientation. Sound has emerged as a leading candidate, because it travels in water irrespective of current flow with little attenuation and because fish and invertebrates create a clamour that can be heard for many kilometers around (4). We have previously shown the attraction of settlement-stage reef fishes from many families to reef noise, using light traps and prerecorded sound (5). Here we provide direct evidence that sound enhances settlement of fish onto patch reefs.

We used two experiments to study settlement behavior in the presence of recorded reef sounds (6). In November 2003, we built 24 patch reefs from dead coral rubble on sand flats in 3- to 6-m-deep water at Lizard Island on the Great Barrier Reef (fig. S1). For six nights, we deployed submersible speakers broadcasting reef noise (at 156 dB relative to 1 μPa at 1 m, mostly the sound of snapping shrimp and fish calls) on 12 of these patch reefs, alternating the location of the speakers each night. Most settlement occurs at night, so recruiting fish were collected from the patch reefs early the following mornings. Of the 868 recruits we collected, most were apogonids (or cardinalfish, 80%) or pomacentrids (or damselfish, 15%). These two families are key members of coral reef fish assemblages around the world: The apogonids contribute up to one quarter of all individuals on reefs and the pomacentrids up to half of the total fish biomass (7). Analyses showed no site or date effects in our data, but both families settled in greater numbers on noisy patch reefs than on silent reefs (Fig. 1A). A preference for noisy patch reefs was also seen in less common fishes, with marginally more taxa (excluding apogonids and pomacentrids) on patch reefs with broadcast noise than on reefs without (Fig. 1B).

Fig. 1.

Comparison of catches from patch reefs with different sound treatments (tables S1 to S3). (A and B) Reefs broadcasting reef noise (black) or silent reefs (white). (C and D) Reefs with high-frequency (black) or low-frequency (gray) reef noise or silent reefs (white). Statistical results are for (A) Chi-squared analyses, (B) Wilcoxon's matched pairs test, (C) pairwise Chi-squared analyses with Bonferroni corrections, and (D) pairwise Wilcoxon's matched pairs test with Bonferroni corrections (ms, P < 0.1; *, P < 0.05; **, P < 0.01). All apogonids and pomacentrids were excluded from the analyses in (B) and (D).

In December 2003, the experimental field site was used to compare the settlement of fishes to patch reefs where we broadcast primarily the high frequencies of reef noise (80% > 570 Hz, predominantly shrimp) or low frequencies of reef noise (80% < 570 Hz, predominantly fish) with settlement to silent reefs. This time, nearly four times as many recruits arrived (3111 fish), but the taxonomic composition was similar. Apogonids settled on high- and low-frequency patch reefs in equivalent numbers, but pomacentrids were preferentially attracted to reefs with high-frequency noise (Fig. 1C). Again, reefs without sound received less settlement from rarer taxa than reefs with broadcast sound (Fig. 1D).

This study provides direct field evidence that settling reef fishes use sounds to orientate toward and select reefs. Furthermore, there is an indication that some fish groups may be selectively using specific components of the reef sound to guide their settlement behavior. The important use of sound at this critical life history phase raises the possibility of potential adverse effects of increasing anthropogenic noise pollution (e.g., shipping and drilling), but it may also lead to the development of new tools for fisheries managers for restocking fisheries or newly established marine reserves.

Supporting Online Material

Materials and Methods

Fig. S1

Tables S1 to S3

References and Notes

View Abstract

Navigate This Article