Timing and Extent of Drift of Shortnose Sturgeon Larvae in the Saint John River, New Brunswick, Canada

Abstract

Little is known about the dispersal of Shortnose Sturgeon Acipenser brevirostrum larvae in the wild. In the Saint John River, New Brunswick, we captured a total of 2,251, 460, 2,100, and 2,083 larvae in 2008–2011, respectively; abundance estimates ranged between 21,000 (2009) and 244,687 larvae (2008). A substantial reduction in larval numbers (49–76%) was recorded over the 4.5-km distance between the two sampling transects deployed in 2008–2010. We found no consistent pattern of larval distribution across the channel, but we recorded a consistent, significant preference for nighttime (dusk to dawn) over daytime dispersal. Generalized linear models were used to examine the timing and extent of larval migration in the Saint John River during the study period. Logistic models incorporating water temperature and Mactaquac Dam discharge provided good predictions of the timing of larval migration. The probability of larval presence was highest when water temperature reached 15°C. At this temperature, larvae were predicted to disperse when nighttime total dam discharge was 20 10⁶ to 30 10⁶ m³. The extent of larval migration was described using negative binomial models, which indicated that dam discharge and transect location significantly influenced the number of drifting larvae. However, data variability was high, reducing predictive capability. Our findings include the first report of Shortnose Sturgeon larval abundances in the Saint John River. The predictions of timing and extent of drift provide information for future sampling and conservation efforts during this vulnerable period as well as insight into the relationships between environmental variables and larval drift in this protected species.

Received April 16, 2012; accepted December 10, 2012

ACKNOWLEDGMENTS

We are grateful to Brent M. Wilson, Laura Qi, Faith M. Penny, Andrew Hazen Brown, Jennifer R. Adams, Christine Adams, and Joel R. Chase for their help with field collections of larvae and image analysis in the laboratory. We also thank the Hartt Island RV Resort for their support during the 2008–2011 sampling seasons. This study was supported by Mathematics of Information Technology and Complex Systems research grants to J.W. and M.K.L., the New Brunswick Wildlife Trust Fund, and Natural Sciences and Engineering Research Council Discovery and Strategic grants to M.K.L. The comments and suggestions provided by the anonymous reviewers greatly improved the manuscript.