A THREE-YEAR STUDY USING LIVE-DEAD ASSEMBLAGES TO ASSESS ANTHROPOGENIC CHANGES IN THE MOLLUSC FAUNA OF A SAND FLAT IN WILMINGTON, NORTH CAROLINA

Human effects on marine ecosystems are increasingly apparent with current events such as the Deepwater Horizon oil spill in the Gulf of Mexico. As part of a three-year Research Experiences for Undergraduates program, samples have been collected yearly to assess the extent of anthropogenic effects in a sand flat near Masonboro Sound, Wilmington, NC. Kidwell (2007) hypothesized that as anthropogenic effects on an environment increased similarities between live and dead assemblages would decrease. Kidwell (2008) also proposed that assemblages collected at the same time would have more similarities than live assemblages collected in successive years. REU students in previous years collected at random 16 two-liter samples from the upper 20 cm of substrate within a 200 m² area of the sand flat. In 2010 we collected another set of samples using these methods to maintain consistency among years. The current year’s assemblages yielded 160 live specimens, including 20 genera (15 bivalve and 5 gastropod genera) and 9559 dead specimens made up of 36 bivalve and 17 gastropod genera. Rarefaction curves comparing the 2010 live and dead bivalve assemblages showed the dead assemblage was significantly richer than the live assemblage. Richness of the current year’s live bivalve assemblage did not differ significantly from that of previous years; the 2010 bivalve dead assemblage also did not differ significantly in richness from previous years. Spearman’s rank correlation coefficient (r_s) supported Kidwell’s hypothesis that live and dead assemblages collected in the same year share more similarities than do live assemblages collected in successive years. Live assemblages of both bivalves and gastropods differed significantly for the three years based upon r_s values. In contrast, the 2010 live and dead bivalve assemblages did not differ significantly (r_s= 0.525). The dead assemblage comparison had significant r_s values (2009 v. 2010 r_s= 0.793; 2008 v. 2010 r_s= 0.666), suggesting similarities in rank-order abundance of genera for the three years. The similarities between live and dead assemblages suggest a lack of anthropogenic effects or may occur because most taxa are rare in the dead assemblage and absent from the live assemblage. The same conclusion was obtained by the earlier REU programs (Dietl et al. 2009, Plaza Muniz et al. 2010).