Genomic signals of local adaptation in UK black alder populations (Alnus glutinosa)

Abstract

Black alder (Alnus glutinosa) is a native tree species in the British Isles, which it colonised after the last glaciation from continental refugia. It is now a widespread pioneer species, thriving on wetlands and riverbanks. We present whole genome sequence data for 324 alder seedlings grown from seed collected from 44 populations across the UK. Among these we found two triploids and one sample with trisomy at chromosome five. Removing these, we found little population genetic structure apart from diverging northern populations with lower diversity. Using Latent-Factor Mixed Models (LFMM) analyses we identified 1,364 loci whose allele frequencies were significantly associated with environmental variables. Of these loci, 93% were associated with the number of snow days. In contrast, the greatest total genomic signal (variance in SNP effect sizes of 7.38 million SNP loci) was observed for mean temperature of the wettest quarter and annual number of frost days. Estimates of maladaptation using 1.49 million SNPs, predicted that northern populations are worst adapted to present and future worst-case climates. Our results suggest that similarity of source-site temperature variables should be prioritised when choosing seed sources for new planting, and northern populations could benefit from assisted gene flow from the south.