Influence of Survey Design on Fish Assemblages: Implications from a Study in Chesapeake Bay Tributaries

Aquatic resource surveys that span multiple decades provide valuable information about fish population responses to environmental and human-induced changes. Often, surveys are altered as scientific objectives change or in response to management needs. These modifications usually result in inconsistencies in the time series, which must be addressed for proper analysis of fish community data. Since 1997, juvenile fishes in Virginia tributaries of Chesapeake Bay have been captured monthly using a bottom trawl at both fixed and random sites. Previous surveys in these tributaries (1955-1996) were conducted at fixed sites only; thus, an understanding of the effect of this design change would allow us to infer fish community status through multiple decades. We compared samples from the fixed site design with those from the stratified random design in the James, York, and Rappahannock rivers and examined species composition, biodiversity, relative abundance estimates, and size distributions to understand the effects of survey design on these metrics. Catches from random sites were characterized by consistently higher species richness than those from fixed sites; biodiversity metrics varied by sampling site type (fixed versus random) and river. For most select species, we observed similar trends in relative abundance regardless of site type; however, for some species we noted differences in magnitude among years and between sampling site types. Community metrics, such as taxonomic diversity and distinctness, revealed subtle differences about fish species assemblages compared with traditional diversity metrics demonstrating that community metrics can characterize biodiversity of fishes at higher taxonomic levels. Thus, species-specific and community metrics derived from assemblage data from fixed-site surveys may not fully represent the magnitude of change in demersal fish assemblages, but can provide reliable indicators of patterns of change in abundance through time. Received July 26, 2012; accepted March 14, 2013