Future Research

It is difficult to evaluate the general well-being of animal populations in the Hudson River estuary and to manage them without knowing how this nursery area is used. This research has provided basic, crucial information about the most abundant species that develop in the Hudson River Estuary and New York Bight during the summer and how their larvae migrate between these areas. Detailed physical-behavioral models have been developed for larvae of each species that collectively show how differential migrations between adult and larval habitats are made. This was accomplished by blending four approaches: 1) a comparative hypothesis testing approach using eggs and larvae, 2) extensive horizontal surveys, 3) intensive hourly sampling over consecutive light-dark and tidal cycles and multiple depths, and 4) comparing larval production to settlement. This sampling strategy has yielded important new insights into the relative contributions of behavior and hydrodynamics to larval transport. Using eggs of different buoyancies as passive tracers indicated that larvae regulated their position in the water column. Contrasts among diverse species with different swimming capabilities and sites with different levels of mixing revealed considerable variation in the abilities of larvae to regulate depth. Comparing vertical migrations by larvae of the same species across the range of habitats encountered during their horizontal migrations revealed the first evidence of tidal vertical migrations and phenotypic plasticity inresponseto changing oceanographic conditions. Contrasting larval abundances and distributions across study sites with known adult distributions revealed the probable consequences of larval behavior, turbulent mixing, and advection to larval transport. Comparing larval production to recruitment revealed that larval export was as reliable as larval retention.

Although our sampling design was a significant advance in studying larval transport of estuarine-dependent species, it was not without its limita tions. Two notable deficiencies were our inability to survey larvae and to monitor recruitment in the lower estuary. Daily monitoring of recruitment over severalmonths is required to gain meaningful data, but it was not feasible to do so from our remote location. A much better understanding of the relationship between larval transport and recruitment could be gained by concurrently monitoring recruits, larval distributions, and physical variables. Our understanding of larval transport processes on the continental shelf also needs to be greatly improved, and I refer the reader to Shanks' (1995) review for an insightful discussion of how to tackle this difficult problem.

Perhaps the biggest gap in our knowledge is how fish larvae use the lower Hudson River Estuary. Even basic informationonlarval distributions andabun-dances is limited, making it difficult to foresee how the complex circulation of the lower bay affects the movements of fish larvae. However, it may be that 1) most larvae migrate to and from the Hudson River Estuary and Raritan Bay along main channels, 2) larvae found in Ambrose Channel likely will be flushed from the lower estuary unless they remain in landward flowing bottom waters, 3) tidal vertical migrations may not greatly facilitate retention or reinvasion over shoals and flanks because flow is seaward throughout the water column, 4) most larvae would recruit during neap tides due to enhanced stratification, 5) eddies may concentrate larvae in calm shoal areas, and 6) flux should be greatest along channels at openings to the lower estuary, depending on winds and tidal amplitude. These hypotheses can be addressed by concurrently describing the circulation, larval distributions, and larval behaviors to estimate larval retention and flux.

In conclusion, we have provided an initial understanding of the relative roles of these two processes in larval depth regulation and transport between the Hudson River Estuary and New York Bight that ultimately may lead to better predictions of year class strength of recreationally, commercially, and ecologically important species. However, an exhaustive sampling strategy should be coupled with promising new approaches, such as individualbased models, larval behavior mimics (Wolcott, 1995) and larval tagging (Anastasia, Morgan, and Fisher, 1998; Thorrold et al., 2002), to fully resolve the roles of behavioral and physical processes in larval transport.

0 0

Post a comment