Vertical swimming behavior influences the dispersal of simulated oyster larvae in a coupled particle-tracking and hydrodynamic model of Chesapeake Bay

Abstract

Because planktonic organisms have swimming speeds that are orders of magnitude lower than horizontal current velocities, it is unclear whether behavior of weak-swimming bivalve larvae could influence dispersal distance, encounters with suitable habitat, or subpopulation connec- tivity. We used a numerical approach to investigate whether these processes could be affected by species-specific differences in larval vertical swimming behavior of 2 oyster species (Crassostrea vir- ginica and C. ariakensis) in Chesapeake Bay, a partially mixed estuary. A coupled particle-tracking and hydrodynamic model was forced with observed winds and freshwater flow and included the best available estimate of present-day oyster habitat. Model scenarios were conducted with hydrody- namic predictions from June to September, 1995 to 1999, to simulate a range of environmental con- ditions. Simple larval swimming behaviors were parameterized for the 2 oyster species with results from preliminary laboratory experiments and literature. To isolate the effect of circulation, settlement habitat, and larval behavior on the spatial trajectories of particles, vertical swimming velocity was the only biological process represented in the model; egg production and larval growth were not included. Differences in larval swimming behavior had significant consequences for particle trans- port in Chesapeake Bay by influencing dispersal distances, transport success, and the degree of con- nectivity between 'subpopulations' in different tributaries. Most particles (>96%) did not return to the same reef on which they were released, and there were behavior-dependent differences in spa- tial patterns of the 'source' and 'sink' characteristics of oyster reefs. Simulated larval behavior had greater influence on spatial patterns of transport success than did interannual differences in circula- tion patterns. These model results have implications for fisheries management and oyster restoration activities.