Subalpine larch (Larix lyallii Parl.) has a narrow ecological niche at timberline in the Cascade Range and the Rocky Mountains of western North America. Demographic factors, including a long generation time (average 500 years) and a late arrival at sexual maturity (100-200 years), make it unlikely that this species will be able to adapt to predicted climate change. A better understanding of genetic structure and genetic diversity is necessary in order to effectively manage this species for future generations. Foliage from 62 populations of subalpine larch was collected in order to elucidate the range-wide population genomics of the species. DNA was extracted and a next-generation sequencing method, restriction site associated DNA sequencing (RAD-seq), was used to generate genome-wide single nucleotide polymorphism (SNP) marker data. Three genetically differentiated clusters were identified via principal components analysis, a discriminant analysis of principal components and Bayesian STRUCTURE analysis: the Cascade Range, the southern Rocky Mountains and the northern Rocky Mountains. A monophyletic group in the central Rocky Mountains was also identified in a dendrogram of genetic distance but this group had weak bootstrap support (49%), meaning genetic differentiation depends on relatively few genetic variants. Genetically differentiated groups should be prioritized for future management and conservation efforts. Negative values of Tajima’s D and preferred demographic scenarios generated by coalescent simulations indicated that 15 populations all have a recent history of expansion. Genetic diversity within these populations was found to be moderate (HO = 0.15 – 0.20), inbreeding coefficients were found to be high (FIS = 0.15 – 0.25) and genetic differentiation among populations was found to be high (average FST = 0.18). These results indicated that fragmentation driven by Holocene warming may have resulted in reduced effective population sizes. Smaller populations experience stronger genetic drift and an increased likelihood of inbreeding, which may hinder an adaptive response to natural selection. Still, parameter estimates for preferred demographic scenarios suggested a minimum effective population size of around 20,000 individuals, which is not considered small by most conservationists. A final study of 18 populations found local adaptation to cold temperature in the northern portion of the species range. In all seasons, populations from the northern Rocky Mountains had significantly higher cold tolerance than populations from the central Canadian Rocky Mountains and the northern Cascades. Winter cold tolerance showed strong clines associated with the frost-free period and degree days below zero. These two climate variables explained 65% of the explainable variance in phenotype when redundancy analysis models were conditioned on geography. Seven SNPs were identified that explained a significant portion of the variance in winter cold tolerance. Range-wide, additional SNPs were identified as FST outliers and/or as significantly correlated with environmental gradients, even after correcting for neutral genetic structure. Together, the results of this work indicate that dispersal, neutral evolutionary processes and natural selection have all played important roles in shaping patterns of genetic variation across the natural range of subalpine larch. All of these factors should be considered during the development of management and conservation strategies for this high-elevation conifer species. / Graduate

Changes within and beyond the alpine treeline ecotone are hypothesized to respond to climatic changes and to be controlled by site-specific conditions. Repeated photographs show significant changes in the alpine treeline ecotone of Goodsir Pass in Kootenay National Park, B.C. over the past century. Field work revealed increases in tree density within the ecotone, and a 150 vertical metre increase in the elevation of the ecotone, at a rate of 2.2 metres/year. Change within the ecotone of Goodsir Pass is more closely related to temporal climatic variability than to site-specific spatial variability. Repeated photographs from three National Parks in the southern Canadian Rocky Mountains show this change to be a typical but dramatic example. Results at several scales indicate that the occurrence, magnitude and type of change in the alpine treeline ecotone and the drivers of that change are most influenced by the regional ecologic and geo-climatic setting or context.