Date of Award

Summer 5-14-2016

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)


Wildlife Ecology and Wildlife Conservation


Joseph Zydlewski

Second Committee Member

Stephen M. Coghlan Jr.

Third Committee Member

Donna L. Parrish


Rainbow Smelt (Osmerus mordax) are an important fish distributed throughout northeastern North America with both anadromous and landlocked populations. Abundance, size at age, and maximum size vary widely among populations and life histories. In order to compare anadromous and landlocked populations, we collected spawning adults in 2014 from four anadromous and three landlocked populations. Scales and otoliths from the anadromous fish were examined and compared for estimates of bias and precision in ageing. Analysis of both scales and otoliths provided age estimates that were acceptable, but estimates from scales were more precise and had less bias. Otoliths were used to estimate mean size at age and von Bertalanffy growth parameters for each population. Compared to landlocked populations, anadromous fish exhibited a greater and more variable size at age, and asymptotic size. While anadromous fish generally grew faster than landlocked fish, von Bertalanffy growth parameters were variable across life histories. Age analysis showed that populations of both anadromous and landlocked Rainbow Smelt were comprised of fish age 1 to 4, and were typically dominated by a single age class. These data suggest considerable plasticity associated with trade offs between growth and reproduction among different populations and life histories.

Commercially reared Rainbow Smelt larvae have recently become available for supplementation for this species that is known to be highly variable in abundance. We stocked smelt larvae into two small ponds in central Maine at a density of approximately 30,000 fish per hectare to assess survival of hatchery reared fish. Fish were double marked with thermal and oxytetracycline marks. We subsequently sampled for stocked larval Rainbow Smelt with ichthyoplankton tows, both day and night for the first four weeks after stocking, capturing more than 1,800 Rainbow smelt in one pond, and two in the other. Capture rate was higher at night than the day, and decreased over the duration of the study. Otoliths were examined from a subset of 339 larval Rainbow Smelt. The median hatch date of all fish was two days after the observed hatch date of our stocked fish. The mean daily growth rate was calculated to vary from a low of 0.3 mm per day at 7 days after hatching, to a high of 0.5 mm per day at 14 days after hatching. There were no distinct marks consistent with oxtretracycline marking found on any of the larval Rainbow Smelt otoliths examined. Potential thermal marks and stocking checks were found on otoliths from 80% and 45% of fish examined respectively. Larval Rainbow Smelt density and distribution was estimated with a linear model and was significantly related to depth, time of day, and sample event. This model was to estimate the population and mortality of Rainbow Smelt. The large difference between the observed and predicted catches of Rainbow Smelt in one of our study waters lends evidence to the poor success of stocking on this water.