Horizontal and vertical trends in the distributions of larval fishes in coastal waters off central New South Wales, Australia

Abstract

Larval fishes were sampled across six transects perpendicular to a 50 km section of the coast off Sydney, Australia, in April/May and August/September 1990. Samples were collected at the surface and at depth (20 to 30 m) at three locations across each transect; over the 30, 70 and 100 m depth contours. There was a large level of heterogeneity in the horizontal and vertical distributions of most taxa examined, and no general pattern of distribution was evident for the whole assemblage. Classification analyses revealed that the major differences between assemblages were related to depth. Horizontal trends in the distributions of the abundant taxa were evident in the inshore-offshore direction, but not longshore. Seven taxa belonging to the families Gobiidae, Labridae, Sillaginidae, Sparidae, Ambassidae, Clupeidae and Clinidae/Tripterygiidae were most abundant inshore, whereas 4 taxa of the families Cepolidae, Percichthyidae, Cheilodactylidae and Gonorynchidae were generally more abundant offshore and 24 taxa showed no discrete horizontal trends across transects. More taxa and individuals were generally caught at depth than at the surface and this was evident across all transects. Twenty taxa were more numerous at depth, whereas 4 taxa, the Cheilodactylidae, Gonorynchidae, Mullidae and Scorpididae, were most abundant at the surface and 11 taxa showed no difference in densities between depths. Ontogenetic differences in the distributions of some larvae were evident. The mean size of larval Liza argentea (Mulgilidae) caught was greater offshore than inshore, and greater at the surface than at depth. In contrast, larger Pseudocaranx dentex (Carangidae) occurred in greater numbers at depth than at the surface. The data emphasise that the assemblages of larval fishes in coastal waters off central New South Wales cannot be modelled as a single unit, which concurs with the findings in other temperate and tropical vaters. Furthermore, the data denote the need to spatially stratify sampling in these waters in order to assess seasonal changes in these assemblages.