The overwintering habitat use of shortnose sturgeon (Acipenser brevirostrum) was investigated from January to March 2005 in the upper Kennebecasis River, New Brunswick, Canada, using a novel underwater video camera system and modeling approach. Following a random sampling procedure, 187 holes were drilled into the ice, and 234 sturgeon were counted and video-recorded. We found that sturgeon concentrated in a 2 ha area at the confluence of the Kennebecasis and Hammond rivers on a flat sandy substrate at a depth of 3.1–6.9 m. Generalized linear models were developed to describe the relationship of shortnose sturgeon density and habitat variables. The model indicated that the shortnose sturgeon had significant preference to deeper areas within this region. The total abundance of shortnose sturgeon in the area was estimated to be 4836 ± 69 (mean ± standard error) using the ordinary kriging method to interpolate sturgeon density at unsampled sites. This overwintering habitat of shortnose sturgeon can be defined as critical habitat following the identification policies of the Canadian Species at Risk Act (SARA).

Larval shortnose sturgeon, reared at 17°C, were subjected to delayed feeding treatments of 0, 5, 10, 15, 18, and 23 days post-yolk absorption to examine effects of food deprivation on growth, survival, swimming activity, and escape capabilities. Starvation affected growth and survival but despite degree of starvation, larvae were able to resume growth and experience high survivorship following feeding. Specific growth rate based on larval dry weight for the period directly following first feeding was highest for the day 15 and 18 delayed feeding treatments. There were no differences in survival between the 0 and 5 day treatments, however survival was reduced to 71.2%, 45.4%, and 28.8% for 10, 15, and 18 day delayed feeding treatments, respectively. Shortnose sturgeon had a point-of-no-return (PNR; 55.7% initiated feeding) at ~19 days (or 42 days post-fertilization) following the full absorption of yolk. Mean percent swimming activity and swimming speeds showed an interaction between delayed feeding treatment and larval age, such that no differences were detected at 1 and 6 days post-yolk absorption, while these swimming behaviors generally increased or spiked as feeding was delayed for 10, 15, and 18 days post-yolk absorption. At 23 days post-yolk absorption, only swimming speed increased for larvae that were denied food for 18 days. While there was an interaction between delayed feeding treatments and age for proportion of larvae exhibiting an escape response, generally, larvae from all feeding treatments exhibited a positive escape response. There were also interactions between delayed feeding treatments and age post-yolk absorption for mean and maximum escape speeds, such that less aggressive escape responses were typically detected the longer larvae were denied food. Our research suggests that larval shortnose sturgeon increase physical activity during periods of starvation to find a food patch while remaining vigilant but maybe not as capable to defend against a predatory attack as fed individuals.