Clam diet and production in relation to the spatial pattern of food source inputs and quality: A stable isotope approach

We used four stable isotopes to determine the spatial variability in the diet of Mya arenaria in the Plum Island Sound estuary, Massachusetts, USA and compared diet to the production rate of dietary food sources. Clams were collected for a year along the estuary and their diet determined at each site with four stable isotopes, δ 13 C, δ 15 N, δ 34 S and δ 2 H, using the Bayesian mixing model (MixSIAR). We compared diet composition and its quality to patterns of clam production. The results showed that M. arenaria depended on locally produced organic matter, the availability of which varied along the estuary. For the lower estuary, the dominant food source for clams was ocean/estuarine unicellular algal particulate organic matter followed by salt marsh detritus. Brackish and riverine organic matter sources were of secondary importance. For the upper estuary, the relative importance of brackish marsh and riverine organic matter increased. The net inputs of organic matter from the river, tidal marshes and unicellular algae were closely matched with the dietary composition of clams. The down estuary pattern of increasing clam productivity was best predicted by a similar pattern in suspended particulate organic matter quality as indicated by the ratio of C:N. The C:N ratio decreased from about 10.5:1 to 7.4:1 along the region of clam growth. We also found that δ 2 H was a useful tracer, effectively discriminating between marsh-derived and aquatic algae-derived organic matters and significantly increasing the predictive power and reliability of the mixing model. Our study provides fundamental knowledge on suitable habitats for the M. arenaria growth in the Plum Island estuary and insights into selecting stable isotopes for elucidating feeding ecology and trophic dynamics of estuarine organisms. • We developed a mixing model for Mya arenaria with δ 13 C, δ 15 N, δ 34 S and δ 2 H. • M. arenaria depended on locally produced organic matter varied along the estuary. • The dietary composition matched the net organic matter inputs from their sources. • The pattern of organic matter quality best predicted the clam productivity pattern. • Adding δ 2 H significantly increased the predictive power of the mixing model.