Understanding how organisms interact, such as in predator-prey systems, is central to the study of animal ecology. In 2004, we initiated a multi-species fish monitoring survey to measure fish abundance, distribution, and species composition in Sarasota Bay using a small purse seine. This work has allowed us to investigate bottlenose dolphin foraging ecology and their role as predators in the ecosystem. Bottlenose dolphins of Sarasota Bay tend to be exclusively piscivorous, foraging alone and feeding on individual fish rather than schools of fish, and they commonly consume prey associated with sea grass habitat. It has been hypothesized that coastal bottlenose dolphins use passive listening to locate noise-producing (i.e., soniferous) prey, however this has never been tested. The purpose of this study was to quantify prey selection for wild, resident bottlenose dolphins in the Sarasota Bay region. This study represents the first attempt to analyze prey selection using the stomach contents of individual cetaceans with known home ranges combined with a robust assessment of the structure of the fish community.
We compared the relative abundances of prey available to estimates of prey use at closely matching spatial and temporal scales. Stomach content analyses by Nélio Barros of 15 dolphins from 1996 to 2006 with extensive sighting histories and well-documented distributions were used to determine prey use. Prey availability in the study area was assessed by a concurrent purse seine survey of the fish community. In total from all five habitats, 477 purse-seine sets were made during the summer months of 2004 through 2007. In total, 208,762 individual fishes, comprising 56 families and 120 species, were documented. We tested for prey selection using three approaches: (1) G-tests, (2) Manly’s index of standardized forage ratios, and (3) Bonferroni simultaneous confidence intervals. Prey selection was determined at the species and family levels, and for soniferous (i.e., noise-making) versus non-soniferous species. While comprising only 6.3% of the total available prey, soniferous fishes accounted for 51.9% of the total prey consumed. G-tests determined that dolphins in this study significantly selected for prey at the species, family, and soniferous/non-soniferous prey levels. Manly’s standardized forage ratios and 95% Bonferroni confidence intervals determined significant positive selection for soniferous prey and significant negative selection against non-soniferous prey. Dolphins selected against several fish families, including Gerridae, Clupeidae, and Sparidae, as well as against all the species within those families.
These data indicate that at the population level resident bottlenose dolphins of Sarasota Bay select soniferous prey. These results lend further support to the hypothesis that bottlenose dolphins use passive listening to locate sound-producing fishes. Passive listening may increase the capture efficiencies of energetically rich prey, as many soniferous fish species increase the frequency and intensity of calls during spawning periods. In addition, this study shows that resident coastal dolphins in Sarasota Bay, on the west coast of Florida, are selective feeders and should not be considered an opportunistic predator. This may be true for coastal bottlenose dolphins throughout the Southeastern United States, in areas with similar habitats and fish communities (i.e., soniferous fishes). Most importantly, prey use by bottlenose dolphins does not necessarily reflect their feeding preferences. Species commonly seen in the diet, such as pinfish, were not selected in proportion to their availability. Likewise, mullet, and mojarras (Gerreidae) are often considered to be important prey, however, our analysis provides no evidence for their selection. The species that were selected by the dolphins in this study occupy a variety of estuarine habitats including seagrass beds, mangrove fringes, sandflats, and the deeper waters of the open bay. Selection data for many of the prey types in this study were associated with high standard errors. These likely resulted from the low sample size of dolphin stomachs analyzed and perhaps from individual variation in feeding strategies and habitat selection. The extent to which individual prey selection in dolphins occurs is unknown, however it is possible that detection and capture of certain species, such as those associated with structures (e.g., gulf toadfish), require learning and experience, and therefore only a subset of the resident dolphins might be able to exploit such prey resources successfully. Future work on variation in prey selection related to season, demography, reproductive condition, and genetic matrilines is needed.
While bottlenose dolphins are federally protected from all forms of harassment, capture, and harvest, various threats still exist. Habitat degradation, overfishing, and harmful algal blooms (i.e., red tide Karenia brevis) may deplete the abundance of soniferous fishes. Anthropogenic noise, such as that caused by power boats, marine construction or demolition, may mask soniferous fish calls and under extreme circumstances could damage dolphin hearing. Each of these threats has the potential to limit the ability of dolphins to locate their preferred prey through passive listening, which may increase the energetic costs of foraging.
This work would not be possible without the help of many dedicated interns and volunteers who have donated their time and effort to this project. We would like to thank NOAA’s Fisheries Service for the primary funding for this project. Harbor Branch Oceanographic Institution’s Protect Wild Dolphins Program and Florida’s State Wildlife Grants Program provided additional funding.