Supplementary methods. Temporal outlier detection.

Figure S1. Heatmaps showing the degree of linkage disequilibrium (r2) between loci in samples with mean posterior membership probability >0.6 to the clusters Iceland inshore (A), East (B), Nuuk (C), and West (D).

Figure S2. Heatmap of pairwise FST values between samples.

Figure S3. Plot of the Bayesian Information Criterion (BIC) for clustering solutions with different numbers of clusters (K) based on all loci (A) and a subset of loci (n = 618) excluding temporal and spatial outliers and loci in high LD (B).

Figure S4. Plots to illustrate the configuration of inferred clustering solutions for K = 2:4 based on all loci (A) and a subset of loci (n = 618) excluding temporal and spatial outliers and loci in high LD (B).

Figure S5. Loading plot representing the contributions of alleles from different linkage groups on the first (A), second (B), and third (C) discriminant function from the DAPC based on the four inferred clusters.

Figure S6. Frequency histograms comparing the distribution of maximum individual posterior membership probabilities generated in the DAPC analysis based on clustering of the ‘pure samples’ (see text; dark blue) and a DAPC analysis based on randomized prior cluster assignment (light blue).

Figure S7. Scatterplot of the mean sample coordinates on the first and second discriminant functions (DF) from a DAPC analysis based on all study samples and four selected reference samples from the Northeast Atlantic previously analyzed in Nielsen et al. (2012).

Figure S8. Summary of results from the Bayesian clustering algorithm structure based on the no-admixture (A and C) and the admixture (B and D) ancestry models, both with correlated allele frequencies and with sampling locations incorporated as priors.

Figure S9. Matrix of pairwise FST (estimated with bayescan) plotted against linkage group position in all cluster comparisons for loci in linkage group 1 (A) and linkage group 7 (B).

Table S1. List of variables initially considered for environmental correlation analysis.

Table S2. Overview of outlier loci identified in the different neutrality tests.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.