Geostatistical Analysis of the Abundance and Winter Distribution Patterns of the Blue Crab Callinectes sapidus in Chesapeake Bay

—Spatial heterogeneity is a striking life history feature of the blue crab Callinectes sapidus in the Chesapeake Bay. Spatial patterns of sexand age-specific habitat use have been well documented and affect the fisheries in the bay. However, a quantitative assessment of the spatial distribution of blue crabs during winter, when they are generally buried in the sediments, has been lacking. We applied geostatistical techniques (variogram modeling and kriging) to 13 years of winter dredge survey (WDS) data to map patterns of blue crab winter abundance. These maps were then quantified to derive a time series of baywide abundance and to examine changing patterns of habitat use and aggregation. Geostatistical abundance estimates were generally similar to those calculated from design-based methods but were more highly correlated with fishery catch per unit of effort. Both abundance time series showed a large and significant decline from 1990 to 2002. Changes in spatial distribution were evaluated using trend maps and a density-weighted centroid. Interannual variation in the latitude of the weighted centroid was positively correlated with baywide abundance, suggesting possible density-dependent changes in distribution. The southward shift in winter distribution at low stock size may increase the vulnerability of blue crabs to exploitation in a winter dredge fishery that occurs only in the southern portion of the bay. Such quantitative and spatially explicit information provides a potentially useful base for constructing population models and evaluating alternative management options. There is increasing recognition of the importance of space and the need for spatially explicit information for understanding the ecology, population dynamics, and management of marine resources (Nishida and Booth 2001). This is not only because of a growing awareness of the risks of not accounting for the spatial distribution of exploited populations (Hutchings 1996; Walters 2003), but also because new methods have enabled space to be incorporated into a variety of stock assessment models (Caddy 1975; Walters et al. 1999; Pelletier et al. 2001). These spatial models and spatially explicit management strategies, such as time-area closures and marine protected areas, are being used increasingly to manage fisheries. Although the goals and data requirements of these modeling and management approaches differ substantially, all require a quantitative description of the distribution of the species in question and most could benefit from an understanding of how that distribution * Corresponding author: [email protected] 1 Present address: University of Wisconsin Center for Limnology, 680 North Park Street, Madison, Wisconsin 53706, USA. Received November 30, 2004; accepted June 29, 2005 Published online November 3, 2005 may change over time or with changes in stock