Tidal shifts in the vertical distribution of bivalve larvae: Vertical advection vs. active behavior

Abstract

Some previous studies, though not all, have reported that vertical distributions of bivalve larvae shift upward in the water column on flood tides and downward on ebb tides. When observed, such shifts have been interpreted as reflecting the vertical migration of larvae (i.e., an active behavioral process). This interpretation assumes that tidally driven vertical advection is insignificant compared to larval swimming speeds, but that assumption may not be valid in all regions. We demonstrated that the vertical distribution of mussel (Mytilus edulis) larvae shifted upward on flood tides on all surveyed dates at sites in the eastern Gulf of Maine. We then used a high-resolution coastal circulation model to predict the vertical distribution of passive particles at one site on four sample dates. The modeled distribution shifts explained 99% of the variance in the larval vertical distribution. Finally, we conducted a multipart meta-analysis of published papers on tidal shifts to assess whether variation in the results among studies could be explained by tidal vertical advection via two proxies—tidal range and shelf slope. For well-replicated studies that could be analyzed with a paired t-test, tidal range explained 78% of the variance in the test statistic. A multiple regression approach applied to distributional shifts from a broader set of studies indicated significant effects of tidal range and interactions between tidal range, slope, and slope heterogeneity. Our results demonstrate the importance of considering passive processes when evaluating how the vertical distribution of weakly swimming meroplankton varies between tidal phases.