Mortality and development of wild Atlantic salmon (Salmo salar) embryos associated with environmental conditions over winter in the Miramichi River basin

Abstract

Dynamic winter conditions in northern streams present considerable challenges for incubating wild Atlantic salmon (Salmo salar) embryos. How these winter conditions affect embryo mortality and development in nature is poorly understood. The goals of the present study were to 1) identify abiotic variables associated with the mortality of wild salmon embryos and 2) assess the accuracy and precision of embryonic developmental prediction methods currently used for conservation and basic research purposes. Over two study years (2013-2014 and 2014-2015), embryos were reared in artificial redds in five active spawning reaches in the Miramichi River basin and mortality was quantified at two sampling events (late March, pre-freshet and late May, post-freshet). Through partial least squares regression analyses, 37.6% of the total variation in embryo mortality was explained by abiotic predictor variables from five continuous measurements of incubation conditions in each reach (water temperature, water level, dissolved oxygen, ice conditions, and substrate characteristics). Each of the five groups of predictor variables contributed uniformly to the explained variation, which suggests that mortality is influenced by multiple interacting abiotic conditions, rather than a single variable. Developmental states based on external morphology were assigned to live embryos and alevins at various points through the incubation periods. Corresponding temperature records were used to calculate predicted developmental progression using three temperature-based relative age methods (cumulative degree-days, effective degree-days, and tau somites). Embryonic development was not observed to match any of the prediction methods. However, a linear model fitting effective degree-days to assigned developmental state was suggested as a first step to improve predictions. Population-specific adjustments to parameters within the effective degree-day or tau somite methods and the inclusion of other abiotic variables are recommended. Results from the present study support a multi-faceted ecological perspective of embryonic mortality and development in nature and may provide a baseline from which to measure the impacts of further environmental change.