Autocorrelation in fish recruitment and environmental data can complicate statistical inference in correlation analyses. To address this problem, researchers often either adjust hypothesis testing procedures (e.g., adjust degrees of freedom) to account for autocorrelation or remove the autocorrelation using prewhitening or first-differencing before analysis. However, the effectiveness of methods that adjust hypothesis testing procedures has not yet been fully explored quantitatively. We therefore compared several adjustment methods via Monte Carlo simulation and found that a modified version of these methods kept Type I error rates near . In contrast, methods that remove autocorrelation control Type I error rates well but may in some circumstances increase Type II error rates (probability of failing to detect some environmental effect) and hence reduce statistical power, in comparison with adjusting the test procedure. Specifically, our Monte Carlo simulations show that prewhitening and especially first-differencing decrease power in the common situations where low-frequency (slowly changing) processes are important sources of covariation in fish recruitment or in environmental variables. Conversely, removing autocorrelation can increase power when low-frequency processes account for only some of the covariation. We therefore recommend that researchers carefully consider the importance of different time scales of variability when analyzing autocorrelated data.

Ninety-eight percent of recently surveyed papers in fisheries and aquatic sciences that did not reject some null hypothesis (H0) failed to report β, the probability of making a type II error (not rejecting H0 when it should have been), or statistical power (1 – β). However, 52% of those papers drew conclusions as if H0 were true. A false H0 could have been missed because of a low-power experiment, caused by small sample size or large sampling variability. Costs of type II errors can be large (for example, for cases that fail to detect harmful effects of some industrial effluent or a significant effect of fishing on stock depletion). Past statistical power analyses show that abundance estimation techniques usually have high β and that only large effects are detectable. I review relationships among β, power, detectable effect size, sample size, and sampling variability. I show how statistical power analysis can help interpret past results and improve designs of future experiments, impact assessments, and management regulations. I make recommendations for researchers and decision makers, including routine application of power analysis, more cautious management, and reversal of the burden of proof to put it on industry, not management agencies.

To improve the understanding of linkages between ocean conditions and salmon productivity, we estimated effects of ocean temperature on survival rates of three species of Pacific salmon (Oncorhynchus spp.) across 120 stocks. This multistock approach permitted more precise estimates of effects than standard single-stock analyses. The estimated effects were opposite in sign between northern and southern stocks and were quite consistent across stocks within species and areas. Warm anomalies in coastal temperatures were associated with increased survival rates for stocks in Alaska and decreased survival rates in Washington and British Columbia, suggesting that different mechanisms determine survival rates in the two areas. Regional-scale sea surface temperatures (SST, within several hundred kilometres of a stock's ocean entry point) were a much better predictor of survival rates than large-scale climate anomalies associated with the Pacific Decadal Oscillation (PDO), suggesting that survival rates are primarily linked to environmental conditions at regional spatial scales. With appropriate cautions, these results may be used to predict the potential effects of climatic changes on salmon productivity in different areas of the Northeast Pacific.Résumé : Dans le but d'améliorer notre compréhension des liens entre les conditions océaniques et la productivité du saumon, nous avons estimé les effets de la température de l'océan et les taux de survie chez 120 stocks de 3 espèces de saumons du Pacifique (Oncorhynchus spp.). Cette approche impliquant de nombreux stocks a fourni des estimations plus précises que l'analyse habituelle de stocks individuels. Les effets estimés sont de signe contraire dans les stocks du nord et du sud et sont uniformes d'un stock à l'autre pour une même espèce dans une même région. Des réchauffements anormaux des températures côtières sont associés à des augmentations des taux de survie des stocks d'Alaska et des diminutions des taux de survie au Washington et en Colombie-Britannique, ce qui laisse croire que des mécanismes différents régissent la survie dans ces deux régions. Les températures de surface de la mer (SST) à l'échelle régionale (sur une distance de plusieurs centaines de km du point d'entrée d'un stock dans l'océan) sont de meilleures variables prédictives du taux de survie que les anomalies climatiques à grande échelle associées à l'Oscillation décennale du Pacifique (PDO), ce qui indique que les taux de survie sont liés principalement aux conditions environnementales à l'échelle spatiale régionale. Avec les précautions appropriées, ces résultats pourraient servir à prédire les effets potentiels des changements climatiques sur la productivité des saumons de différentes régions du nord-est du Pacifique.[Traduit par la Rédaction] Mueter et al. 463

No abstract

We examined spatial correlations for three coastal variables [upwelling index, sea surface temperature (SST), and sea surface salinity (SSS)] that might affect juvenile salmon (Oncorhynchus spp.) during their early marine life. Observed correlation patterns in environmental variables were compared with those in survival rates of pink (O. gorbuscha), chum (O. keta), and sockeye (O. nerka) salmon stocks to help identify appropriate variables to include in models of salmon productivity. Both the upwelling index and coastal SST were characterized by strong positive correlations at short distances, which declined slowly with distance in the winter months, but much more rapidly in the summer. The SSS had much weaker and more variable correlations at all distances throughout the year. The distance at which stations were no longer correlated (spatial decorrelation scale) was largest for the upwelling index (> 1000 km), intermediate for SST (400–800 km in summer), and shortest for SSS (< 400 km). Survival rate indices of salmon showed moderate positive correlations among adjacent stocks that decreased to zero at larger distances. Spatial decorrelation scales ranged from approximately 500 km for sockeye salmon to approximately 1000 km for chum salmon. We conclude that variability in the coastal marine environment during summer, as well as variability in salmon survival rates, are dominated by regional scale variability of several hundred to 1000 km. The correlation scale for SST in the summer most closely matched the observed correlation scales for survival rates of salmon, suggesting that regional‐scale variations in coastal SST can help explain the observed regional‐scale covariation in survival rates among salmon stocks.