Migration and spawning of anadromous shads in the Roanoke River, North Carolina

Anadromous alosines are ecologically, commercially, and recreationally valuable fishes. Some populations of alosines are at historically low levels, as a result of overfishing, pollution and habitat change, including the presence of dams, which block access to historic spawning sites upstream and alter spawning habitat downstream. To aid in the restoration of alosine stocks, I evaluated several methods for spawning habitat identification and characterization and evaluated trap and transport as a method to give American shad Alosa sapidissima access to additional spawning habitat above dams in the Roanoke River, North Carolina and Virginia. The Roanoke River has four anadromous alosines: American shad, hickory shad A. mediocris, blueback herring A. aestivalis, and alewife A. pseudoharengus. For all four alosines, spawning sites river-wide were most efficiently identified using plankton tows, which collect eggs of all species. Spawning habitat selection for hickory shad and river herring (blueback herring and alewife) could be evaluated using spawning pads, which collected their eggs in clumped distributions. American shad eggs were not successfully sampled by spawning pads and spawning habitats could be best examined by visual observations of spawning splashes. Unlike the other alosines in this study, little is known about spawning habitat for hickory shad. Hickory shad eggs were collected at water temperatures from 10 to 23°C, and peaked from 11 to 14°C. Spawning generally occurred in water velocities ≥ 0.1 m/sec, over substrates free from silt. A habitat suitability model for hickory shad was developed using a Bayesian belief network. Bayesian belief networks are a relatively new method for modeling habitat suitability for fishes, but could prove very useful in the future, especially for species such as American shad which have been the focus of more study and are in need of restoration. I evaluated movement patterns and spawning of sonic-tagged adult American shad transported to habitats above dams on the Roanoke River. Most transported fish spent relatively little time in the riverine habitat considered suitable for spawning, and no eggs were collected by plankton sampling. American shad appeared to move more effectively through a smaller, as compared to a larger, reservoir, but fish released directly into riverine habitat spent the longest amount of time in suitable spawning habitat. Although the mortality associated with moving downstream through a dam turbine was generally low, few adults completed the passage, and many were observed just upstream from a dam late in the season, suggesting that structures to increase downstream passage may be beneficial. I used data on behavior and outmigration of American shad adults and fry released above and below dams on the Roanoke River to develop a deterministic, density-dependent, stage-based matrix model to predict possible population-level effects of transporting American shad to upper basin habitats. The American shad population in the Roanoke River appears small compared to assumed values of carrying capacity in the lower river and would appear to benefit from transport only under optimal conditions of young survival and effective fecundity. The matrix model predicted that under present conditions, improvements to survival rates of young or adults would likely lead to greater improvements in the stock size of American shad in the Roanoke River.