Pilot Habitat Assessment of a Mesohaline Embayment of the Chesapeake Bay

Acoustic habitat mapping and fish census techniques were integrated in an attempt to quantify the ecological importance of oyster shell habitats in the Chesapeake Bay. Forty-one km² of seabed was mapped at the confluence of the Rhode, South, and West rivers in Maryland. A side scan sonar system provided two dimensional textural imagery and a single beam echosounder collected bathymetric and seabed classification data. Cover maps were derived from the integration of sidescan sonar, acoustic classification and grab sample data. Benthic habitat classifications were Mud, Sand, Mud/Sand, Patchy 3D Oyster Shell with Mud, 2D Oyster Shell with Mud, 2D Oyster Shell with Sand. An otter trawl was used to collect organisms present within GIS derived habitat polygons. Seventy-five samples were collected. Multibeam sonar was used to collect high resolution bathymetry data over trawl transects. Slope and rugosity grids were created from multibeam bathymetry. We used generalized linear models to assess the relationship between fish community metrics and a suite of habitat variables. Dependent variables were abundance of pooled fish species (no./m²), the Shannon-Wiener diversity index, and abundance of the five most frequently observed fish taxa. Independent variables were benthic habitat type, season, a habitat type x season interaction, bottom salinity, bottom DO, bottom temperature, and trawl start depth.

Benthic habitat type, followed by the habitat type x season interaction term, were the most significant factors contributing to variation in fish community metrics. One-way analyses identified significant variation in pooled abundance and individual species abundance relative to habitat type alone. Diversity did not vary significantly with habitat type. Contrary to expectations, observed abundance was generally greatest on Clay/Silt, Silt/Sand, and Oyster Shell with Mud bottoms; abundance was lowest on Sand bottoms. Preliminary analysis of multibeam data did not reveal strong relationships between abundance and slope or rugosity due to high leverage induced by outliers. These data suggest, however, that trawl efficiency is depressed in highly rugose oyster shell habitats.