Many organisms experience fasting in their life time, and this physiological process has the potential to alter stable isotope values of organisms, and confound interpretation of food web studies. However, previous studies on the effects of fasting and starvation on stable isotopes show disparate results, and have never been quantitatively synthesized. We performed a laboratory experiment and meta‐analysis to determine how stable isotopes of δ 15N and δ 13C change with fasting, and we tested whether moderators such as taxa and tissue explain residual variation. We collected literature data from a wide variety of taxa and tissues. We surveyed over 2000 papers, and of these, 26 met our selection criteria, resulting in 51 data points for δ 15N, and 43 data points for δ 13C. We determine that fasting causes an average increase in the isotopic value of organisms of 0.5‰ for δ 15N and that the only significant moderator is tissue type. We find that the overall effect size for δ 13C is not significant, but when the significant moderator of tissue is considered, significant increases in blood and whole organisms are seen with fasting. Our results show that across tissues and taxa, the nutritional status of an organism must be considered when interpreting stable isotope data, as fasting can cause large differences in stable isotope values that would be otherwise attributed to other factors.

Trophic interactions within and among species vary widely across spatial scales and species' ontogeny. However, the drivers and implications of this variability are not well understood. Juvenile Chinook salmon Oncorhynchus tshawytscha have a wide distribution, ranging from northern California to the eastern Bering Sea in North America, but it is largely unknown how their feeding ecology varies and changes with ontogeny across this range. We collected juvenile Chinook salmon and zooplankton using standardized protocols along the coastal Northeast Pacific Ocean. Using a combination of stomach contents and stable isotopes of nitrogen (δ 15 N) and carbon (δ 13 C) to characterize feeding ecology, we found regional differences in prey utilization by juvenile Chinook salmon. With growth and ontogeny, juvenile salmon in all regions became equilibrated with oceanic isotopic values. There were regional differences in the δ 13 C values of juvenile Chinook salmon that may correspond to regional differences in sea surface temperature. There were also regional differences in stable isotope-derived trophic level, and these estimates differed from those derived from stomach contents, possibly due to the different periods over which these metrics integrate. Dietary niche width, as indicated by stable isotopes, corresponded to the expected dietary diversity from stomach contents, combined with the isotopic variability seen in baseline values. Our results indicate strong geographic and ontogenetic differences in feeding ecology of juvenile Chinook salmon. These differences are likely influenced by a combination of ocean-entry date, ocean-entry size, ontogeny, growth rates and regional conditions.

Summary1. Ontogenetic niche shifts are widely prevalent in nature and are important in shaping the structure and dynamics of ecosystems. Stable isotope analysis is a powerful tool to assess these shifts, with d 15 N providing a measure of trophic level and d 13 C a measure of energy source.2. Previous applications of stable isotopes to study ontogenetic niche shifts have not considered the appreciable time lag between diet and consumer tissue associated with isotopic turnover. These time lags introduce significant complexity into field studies of ontogenetic niche shifts.3. Juvenile Chinook salmon (Oncorhynchus tshawytscha) migrate from freshwater to marine ecosystems and shift their diet from feeding primarily on invertebrates to feeding primarily on fish. This dual ontogenetic habitat and diet shift, in addition to the long time lag associated with isotopic turnover, suggests that there is potential for a disconnect between the prey sources that juvenile salmon are consuming, and the inferred prey sources from stable isotopes. 4. We developed a model that considered ontogenetic niche shifts and time lags associated with isotopic turnover, and compared this 'ontogeny' model to one that considered only isotopic turnover. We used a Bayesian framework to explicitly account for parameter uncertainty. 5. Data showed overwhelming support for the ontogeny model relative to the isotopic turnover model. Estimated variables from best model fits indicate that the ontogeny model predicts a much greater reliance on fish prey than does the stomach content data. Overall, we found that this method of quantifying ontogenetic niche shifts effectively accounted for both isotopic turnover and ontogenetic diet shifts; a finding that could be widely applicable to a variety of systems.

Winter is thought to be a critical period for many fish in the ocean, but their ecology during this time tends to be poorly understood. We quantified the feeding ecology of juvenile Chinook salmon (Oncorhynchus tshawytscha) off the west coast of Vancouver Island in British Columbia, Canada, in autumn and winter to determine how seasonality could affect diet. Using stomach contents and stable isotopes, we tested the hypothesis that the winter diet of juvenile Chinook salmon differs from that of the autumn diet. Stomach-content data showed a shift from a primary reliance on amphipods in autumn to euphausiids in winter. This finding was generally corroborated by the stable isotope analysis, although mixing models suggested a greater contribution of fish prey to the diet in both autumn and winter. Understanding the diet of fish during winter may provide useful information for management as a first step in understanding the factors influencing mortality across life stages.

Current investment in conservation is insufficient to adequately protect and recover all ecosystems and species. The challenge of allocating limited funds is acute for Pacific salmon Oncorhynchus spp. in Canada, which lack a strategic approach to ensure that resources are spent on actions most likely to cost‐effectively recover diminished populations. We applied the Priority Threat Management framework to prioritize strategies most likely to maximize the number of thriving Pacific salmon populations on the Central Coast of British Columbia, Canada. These included 79 genetically, ecologically and spatially distinct population groups called conservation units (CUs) for five salmon species. This region has high salmon biodiversity and spans the territories of four First Nations: the Heiltsuk, Nuxalk, Kitasoo/Xai'xais and Wuikinuxv. Using structured expert elicitation of Indigenous and other experts, we quantified the estimated benefits, costs and feasibility of implementing 10 strategies. Under a business‐as‐usual scenario (i.e. no additional investments in salmon conservation or management), experts predicted that only one in four CUs would have >50% chance of achieving a thriving status within 20 years. Limiting future industrial development in salmon habitats, which was predicted to safeguard CUs from future declines, was identified as the most cost‐effective strategy. Investment in three strategies: (a) removal of artificial barriers to fish migration, (b) watershed protection and (c) stream restoration—at 11.3M CAD per year—was predicted to result in nearly half (34 of 79) of the CUs having a >60% chance of meeting the conservation objective. If all conservation strategies were implemented, experts estimated a >50% probability of achieving a thriving status for 78 of 79 CUs, at an annual cost of 17.3M CAD. However, even with the implementation of all strategies, most sockeye salmon CUs were unlikely to achieve higher probability targets of reaching the objective. Policy implications. We illustrate how Priority Threat Management can incorporate the perspectives and expertise of Indigenous peoples and other experts to prioritize conservation strategies based on their cost, benefit and feasibility. Implementation of this framework can help safeguard and recover Pacific salmon in Canada, and could also be used to prioritize actions for other conservation issues globally.