Habitat selection of two gobies (Microgobius gulosus, Gobiosoma robustum):: influence of structural complexity, competitive interactions, and presence of a predator

Herein I compare the relative importance of preference for structurally complex habitat against avoidance of competitors and predators in two benthic fishes common in the Gulf of Mexico. The code goby Gobiosoma robustum Ginsburg and clown goby Microgobius gulosus (Girard) are common, ecologically similar fishes found throughout the Gulf of Mexico and in the southeastern Atlantic Ocean. In Florida Bay, these fishes exhibit habitat partitioning: G. robustum is most abundant in seagrass-dominated areas while M. gulosus is most abundant in sparsely vegetated habitats. In a small-scale field survey, I documented the microhabitat use of these species where their distributions overlap. In a series of laboratory experiments, I presented each species with structured (artificial seagrass) versus nonstructured (bare sand) habitats and measured their frequency of choosing either habitat type. I then examined the use of structured versus nonstructured habitats when the two species were placed together in a mixed group. Finally, I placed a predator (Opsanus beta) in the experimental aquaria to determine how its presence influenced habitat selection. In the field, G. robustum was more abundant in seagrass and M. gulosus was more abundant in bare mud. In the laboratory, both species selected grass over sand in allopatry. However, in sympatry, M gulosus occupied sand more often when paired with G. robustum than when alone. G. robustum appears to directly influence the habitat choice of M gulosus: It seems that M gulosus is pushed out of the structured habitat that is the preferred habitat of G. robustum. Thus, competition appears to modify the habitat selection of these species when they occur in sympatry. Additionally, the presence of the toadfish was a sufficient stimulus to provoke both M. gulosus and G. robustum to increase their selection for sand (compared to single-species treatments). Distribution patterns of M gulosus dand G. robustum likely result from a synthesis of various biotic and abiotic filters, including physiological tolerances to environmental factors, dispersal ability of larvae, and availability of food. Selection for structural complexity, competition, and presence of predators may further define the resulting pattern of distribution observed in the field. Published by Elsevier Science B.V.