Oxygen stable isotope analysis of fish otoliths from the genus Salvelinus: Preservation effects, fractionation, and latitudinal variation

For cold adapted species such as Arctic chary (Salvelinus alpinus , L.), climate-warming impacts will be variable throughout their range and particularly negative at the southern extremes of their range. Accordingly, there is an urgent need to improve understanding of climate-related variability in the biological performance of Arctic chary stocks in order to develop appropriate management actions to mitigate possible negative climate change effects. Otolith oxygen isotope analysis overcomes existing information deficiencies by facilitating reconstruction of the environmental history of individual fish through the use of temperature-dependent fractionation equations. Utilizing otolith S 18O isotopes, the current study objectives were: (1) to examine potential preservation effects on otolith derived delta18O-fractionation, (2) to develop a Salvelinus-specific delta 18O-fractionation equation and, (3) to determine whether evidence exists for latitudinal variation in growth across the Canadian Arctic and sub-Arctic distribution of the species that is consistent with the countergradient variation hypothesis.;Combination of field collected otoliths and monitored water temperatures from sites across the northeastern Canadian distribution of Arctic charr and brook charr (Salvelinus fontinalis, Mitchill) facilitated the estimation of a statistically significant delta18O-fractionation equation (F = 164.588, P < 0.05, r2 = 0.93):; 10001na=17.68˙103T -1K-30.743;When compared to literature reported equations, the fractionation equation developed here was found to share a common slope (F = 0.11, df = 2,108, P > 0.05) but differed in its intercept (F = 83.33, df = 2,110, P < 0.05). Similarities among slope estimates suggest a common otolith delta18O incorporation response among species that may be interpreted as widespread equilibrium otolith delta18O deposition. Dissimilarities among intercept estimates question broad applicability of any single fractionation equation to all species and are associated with significant predictive biases when fractionation equations are used to infer residential water temperatures for species other than those used in the development of the equation. Results do not agree with conclusions of earlier literature studies concerning the wide applicability of fractionation equations to diverse taxa or the derivation of temperatures from those equations that are in close agreement with pertinent temperature data.;The use of otolith chemistry as a tool for inferring the ecology of fishes implicitly assumes that handling and preservation procedures do not alter the isotopic composition of the otolith. Storage and handling is known to affect measured metal concentrations. However, no similar information is available on possible delta18O isotope alterations arising from standard preservation procedures. To test for potential preservation effects, sagittal otoliths from 30 brook charr (Salvelinus fontinalis, Mitchill), and Atlantic salmon smolts (Salmo salar, L.) were removed and examined in replicate treatment and control experiments. For each specimen a single otolith was removed as a control and analysed for delta 18O. The remaining otolith was left in the cranial cavity of the preserved fish. Specimens were randomly assigned to one of six preservation media-temperature storage treatments for 120 days (ethanol: hot, ambient, and cold; formalin: hot, ambient, and cold), after which the remaining otolith was removed, analysed for delta18O and compared to control values using a 3-way (temperature, species, preservative) ANOVA. Of the seven possible single factor and interaction effects, only the species and preservative-species interaction effects were significant (P<0.05). Possible causes for observed species and species-preservative effects are discussed in relation to species-specific differences in otolith chemistry related to growth and environment including the chemical mechanisms of: (1) dissolution-recrystallization involving the precipitation of secondary minerals within and at the otolith surface, (2) adsorption of ions at available binding sites on the otolith surface, and, (3) isotopic exchange during otolith surface dissolution and/or reprecipitation processes, that may act in concert with biologically driven differences in otolith chemistry. Biologically, study findings have implications for the use of preserved otoliths for paleoclimatic reconstructions and the development of delta18O-fractionation equations for thermal history studies.;The relationship between latitude, growth rate, and thermal opportunity for growth (TOG; calculated as the sum of the degree days greater than 2.57°C) was estimated for young-of-the-year Arctic chart data available for nine populations distributed across the Canadian Arctic and sub-Arctic. Evidence for latitudinal variation in growth consistent with the countergradient variation hypothesis was examined. Linear regression analysis determined a significant negative relationship between TOG (regression F = 351.973, r2 = 0.79, P<0.001) and latitude and a significant positive relationship between growth rate (mm/TOG) and latitude (regression F = 451.373, r2 = 0.83, P<0.001). Similarly, evidence for latitudinal gradients in otolith-oxygen isotope measures potentially indicative of changes in physiological function with latitude were examined. A total of 96 sagittal otoliths were run for oxygen isotope analyses. Statistical comparison of regression lines for otolith and water delta18O values, respectively, regressed against latitude found significantly different slopes (t = 2.45, P = 0.029). Results suggest that metabolism may affect otolith-oxygen isotope fractionation. Otolith delta 13C values regressed against otolith delta18O values indicated three distinct groupings of data corresponding to the three major latitudinal clusters from which samples were obtained for study. Separate linear regression analyses on the latitudinal groupings yielded significant positive relationships (P<0.001) in all but Lake Hazen (P = 0.931), indicating shifts in metabolism alone across latitudes cannot account for the high growth rates observed at northern latitudes.