The goal of my PhD research with Dr. Ryan Danby was to characterize and explain landscape-scale variability in alpine treeline characteristics. Treeline positions are controlled by air temperature at broad scales, so models of future vegetation change predict that treelines will advance upslope and northwards as climate warms. This could mean that tundra habitat will be overrun by boreal forest in the near future. However, past treeline responses to warming have been highly variable, particularly in alpine regions, suggesting that topography can strongly influence these dynamics.
Methods and Results:
To address the uncertainty surrounding alpine treeline responses to warming, we used vegetation surveys, dendrochronology, and remote sensing to characterize variability in plant community composition, tree and shrub growth rates, and tree spatial patterns across treelines on different slope aspects and angles in the Kluane Region of southwest Yukon. We then attempted to explain this variability using measured environmental and climatic variables.
We found that treeline plant communities varied far more between slope aspects than between forest and tundra elevations, mainly because active layers are significantly deeper and soil temperatures are significantly warmer on south aspects than they are on north aspects. This suggests plant communities may not advance upslope in synchrony with warming air temperatures. Instead, we might see some of the more competitive plant species on south aspects start to colonize north aspects once the permafrost on north-facing slopes thaws and soil temperatures begin to rise. You can read more about this in our 2017 Journal of Vegetation Science paper on the Publications and Presentations page.
Interestingly, tree and shrub growth (ring width) patterns did not vary between aspects (though ring widths were much wider on south aspects overall). Instead, patterns varied markedly between mountain ranges with different slope angles. We also found that tree growth stopped responding positively to warm summer temperatures after the 1970s. Instead, warm spring temperatures began to slow growth, likely because warm springs deplete the snowpack, which, in turn, decreases the amount of meltwater available to trees and shrubs early in the growing season. This phenomenon was particularly prevalent in mountain ranges with steeper slopes, suggesting future growth of the dominant woody species in these ranges might be negatively rather than positively influenced by climate warming. These findings were divided into two manuscripts, one of which was recently published in Climatic Change (tree growth only) and the other in Arctic, Antarctic and Alpine Research (tree-shrub growth comparison). More info on these articles can be found on the Publications and Presentations page.
Finally, we found that tree spatial patterns differed much more between slope aspects than between elevations or mountain ranges. Tree stems were highly clustered on south aspects, and treeline ecotones were significantly more abrupt than they were on north aspects. These results suggest that treelines on south aspects are less likely to advance in synchrony with climate warming than those on north aspects, since clustered stems and abrupt ecotones are indicative of barriers to seedling establishment above treeline. In the Kluane Region, we suspect these barriers occur due to competitive interactions between tall deciduous shrubs and tree seedlings. Shrub cover on south aspects is already very high and will likely continue to increase as climate warms. So even if climatic conditions above treeline become suitable for adult trees, any tree seedlings that germinate there will likely be out-competed by tall shrubs before they can grow into trees. These findings have been submitted to Annals of the American Association of Geographers.
All in all, my PhD thesis demonstrates that plant responses to warming in alpine regions are likely to vary markedly between slope aspects and angles due to differences in soil temperature and moisture, respectively, and that models of future vegetation change should account for this whenever possible.