Linking juvenile fish and their habitats: An example from Narragansett Bay, Rhode Island

We used two methods and existing field survey data to link juvenile fish an d their habitats. The first method used seine survey data collected monthly from July to October 1988-1996 at fixed stations in Narragansett Bay, Rhod e Island. Thirteen fish species making up 1% or more of the catch were anal yzed by principal components analysis for two time periods: July-August and September-October. The stations were then plotted by their principal compo nent scores to identify station groupings and habitat types. The second met hod used environmental data collected in July and August 1996 at the establ ished survey stations in a principal components analysis. The stations and 13 most abundant species were plotted by principal components scores result ing from the environmental data. For the environmental data, the first two principal components explained 59% of the variance. The first principal com ponent described the amount of energy shaping the habitat and was positivel y correlated with salinity, dissolved oxygen, current flow, and slope, and negatively correlated with silt. The second component was positively correl ated with depth and silt, and negatively correlated with dissolved oxygen. The environmental data grouped the stations according to their distance fro m the ocean and three habitat types emerged. The uppermost station was a si lty barren having low salinities and dissolved oxygen. Three other stations grouped together as low energy, protected habitats with sandy substrates. Lower bay stations had higher salinities, higher dissolved oxygen, higher f low rates, greater slopes, and larger size substrates, mostly cobble and gr avel. Results from the fish data grouped the stations similarly. Combining results from both datasets revealed the uppermost station had the highest c atches, most species, and greatest number of winter flounder (Pseudopleuron ectes americanus) juveniles. Plots of winter flounder catches with principa l component scores from the environmental data indicated the winter flounde r distribution in the bay has shrunk from baywide to mostly the upper estua ry near their primary spawning grounds. Results illustrate the value of cou pling historic fish survey data with environmental measurements for identif ying previously undervalued habitats important to fish.