Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.


Figure S1. Root mean square error (RMSE) of parameter estimation against number of partial least squares included under four demographic models: (a) isolation-by-distance (IBD), (b) contemporary ecological niche model (cENM), and (c) dynamic ENM (dENM).

Table S1. Geographic locations of sampled individuals and their assigned population (see Fig. 1 for distributional details).

Table S2. List of nuclear loci sequenced in this study.

Table S3. Settings for NGen sequence assembler (DNASTAR) used for the 454 data set in the discovery of polymorphic loci.

Table S4. Length of each locus and sampling per populations for each locus.

Table S5. Soil properties used in the construction of soil layers for the principal component analysis analyses (for detailed description see McKenzie et al. 2000).

Table S6. Molecular indices calculated per locus and presented for each population separately (see Fig. 1 for distributional information), as well as across all populations, including heterozygosity (H) and the standard deviation (H_sd), the number of segregating sites (S), the number of haplotypes (K), and nucleotide diversity (π).

Table S7. List of summary statistics used in approximate Bayesian computation analyses.

Table S8. Pairwise FST of the six populations ordered from north to south (lower triangle) and the significance (upper triangle).

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.