Landscape Genomics of the Tussock Cottongrass (Eriophorum vaginatum) and the Dwarf Birch (Betula nana) in North Central Alaska

Elizabeth Anne Stunz, University of Texas at El Paso


Global climate change has resulted in geographic range shifts of flora and fauna at a global scale. Extreme environments, like the Arctic, are seeing some of the most pronounced changes. This region covers 14% of the Earth’s land area, and while many arctic species are widespread, understanding ecotypic variation at the genomic level will be important for elucidating how range shifts will affect ecological processes. Increase in shrub cover is a major effect of ongoing climate change in arctic tundra ecosystems. The relative increases in abundance and cover of shrub species such as birch, willow, and alder (Betula, Salix, and Alnus spp.) are predicted to modify ecological communities by altering ecosystem processes and outcompeting other arctic plant species, such as the tussock cottongrass (Eriophorum vaginatum L.). Eriophorum vaginatum is a foundation species of the moist acidic tundra, whose potential decline due to competition from shrubs may affect ecosystem stability in the Arctic. Here, I examine the genomic population structure, local adaption, and genotype-environment associations of two widespread arctic plant species, the tussock cottongrass(Eriophorum vaginatum) and the dwarf birch (Betula nana L.) using thousands of genomic markers obtained from double-digest Restriction-site Associated DNA sequencing (ddRAD-seq). I then compare environmental niche models for both species from the Last Glacial Maximum (LGM) to the year 2070 to examine the potential of range expansion and persistence of each species in a warming arctic. In Chapter 1, genetic variation was identified in 273 individuals of E. vaginatum from 17 sites along a latitudinal gradient in north central Alaska. These 17 sites were selected due to their inclusion in 30+ years of ecological research as well as their location within a region that was part of the Beringian refugium. A genome-wide SNP dataset was used to investigate population structure, genomic diversity, genotype by environment association and environmental niche modeling. A comprehensive dataset of 3,879 loci and 10,734 SNPs was used to conduct genotype by environment association analyses and revealed environmentally-associated variation. A neutral dataset of 2,776 loci represented by a single SNP was used to conduct population structure, genomic diversity and landscape resistance analyses across the sampled range of E. vaginatum, and supported multiple genetic clusters across sites, including a genetic break between populations north and south of treeline. Gene flow, landscape resistance, and genotype-environment association analyses all supported the influence of subrefugial isolation, contemporary isolation by resistance, and adaptation on current population structure. Using genotype-environment association analyses, 45 candidate loci were identified, with most identified genesrelated to abiotic stress. Our results supported a hypothesis of limited gene flow related to both spatial and environmental factors for E. vaginatum. These results, in combination with life history traits, suggest a limited range expansion of southern ecotypes northward as the tundra warms. These results also have implications for northern ecotypes, as lower competitive attributes may put this foundational species at a disadvantage as the tundra warms and shrub cover increases. In Chapter 2, I used a genome-wide SNP dataset to investigate population structure, genomic variation, and local adaptation of 109 B. nana individuals from 9 sites along a latitudinal gradient in north central Alaska. These sites were chosen to overlap with those sampled for E. vaginatum in Chapter 1 to allow for a comparison of population structure, genomic variation, and adaptation of the two species in the same region. A neutral SNP dataset of 1,039 loci (each represented by a single SNP) was used to demonstrate two genetic clusters, one composed of individuals from the No Name (NN) site and the other composed of individuals from all other sites. The general lack of population structure and absence of allelic variation related to environment along the cline of the latitudinal gradient was likely due to high co-ancestry, incomplete lineage sorting of a relatively continuous population with recent isolation, or previously disjunct populations reconnected by contemporary widespread gene flow. The low levels of co-ancestry of NN with the other sites in addition to the high number of private alleles identified for NN may indicate the presence of B. glandulosa, or an admixed variant between B. nana and B. glandulosa at this site. The lack of structure related to environment, treeline, and geography suggests that B. nana did not share a similar evolutionary history with E. vaginatum along the same latitudinal gradient. The increased prevalence of Betula pollen and macrofossils in the region during warming fluxes during the early and mid-Holocene, and generally higher levels of Betula pollen south of the Brooks Range further suggest that while Betula, and potentially B. nana, was present on the north side of Brooks Range during the LGM, the genus was likely not prevalent. Post-glacial expansion of southern B. nana populations northward could also lead to high levels of co-ancestry between populations north and south of the Brooks Range. The lack of genetic structure and genetic signature related to environmental variation could indicate a higher tolerance for environmental shifts (plasticity) across the range, which could facilitate a competitive advantage for genotypes under climate change. In Chapter 3, I conducted a literature review of graminoid and deciduous shrub distribution patterns in the Arctic and used environmental niche modeling to investigate temporal variation of suitable habitat for E. vaginatum and B. nana in Alaska. Environmental niche models (ENMs) demonstrated small areas of moderately suitable habitat for B. nana during the LGM, with general increases in suitable habitat area through the Mid-Holocene and the present, primarily in southern and west central Alaska. For E. vaginatum, highly suitable habitat decreased in cover from the LGM to the Mid-Holocene, and both moderately and highly suitable habitat continued to decrease in the present, with a suitable habitat shift northward since the LGM. Importantly, and as supported by the literature review, modeling actual species distributions is complex, and the incorporation of population-level and ecological community factors would improve predictions of realistic expansion and population persistence under climate change. While shrubs are expected to increase in height and density, and landscape resistance could hinder range expansion of graminoids as the Arctic warms, fine-scale environmental variation, nutrient availability, dispersal, gene flow, genetic differentiation, local adaptation, competition, and community structure will also shape both species distributions. While B. nana will likely expand rapidly in areas of highly suitable habitat in Alaska, as supported by a lack of local adaptation and narrow environmental tolerances (Chapter 2), moist and warm sites will likely see the greatest increases in density. While E. vaginatum may be able to dominate in colder and/or wetter sites, especially as deeper rooting will allow access to nutrients in thawed permafrost layers, E. vaginatum extent is not expected to increase in the taiga, where landscape resistance is high, treeline reduces gene flow between populations, and competition with shrubs will likely limit population expansion in the warming arctic tundra. My dissertation work utilizes genome-wide SNP datasets to investigate the evolutionary potential of two arctic plant species under climate change. Using landscape-level genomic analyses, population structure, genotype-environment associations, and gene flow barriers were well-supported for E. vaginatum and lacking for B. nana, highlighting disparate evolutionary histories and trajectories for ...

Subject Area


Recommended Citation

Stunz, Elizabeth Anne, "Landscape Genomics of the Tussock Cottongrass (Eriophorum vaginatum) and the Dwarf Birch (Betula nana) in North Central Alaska" (2022). ETD Collection for University of Texas, El Paso. AAI30242395.