Lake-mediated climate buffering in Canada’s western boreal forest

As climate change affects Canada’s boreal forest, cold-adapted species face threats from warming temperatures and competition from encroaching tree species, particularly on the south boundary where some degree of forest compositional change occurs. Climate-change refugia, or areas projected to experience the effects of climate change more gradually, can serve as remnant suitable habitat or as stepping stones for species to disperse in latitude and altitude. Large, deep lakes can act as climate-change refugia along their shorelines, as cold water brought to the surface during annual turnover or upwellings result in the cooling of shoreline temperatures. The largest lake in the world by surface area, Lake Superior, serves as a model system for lake-mediated temperature buffering, as its cool water temperatures and wave action have maintained shoreline habitats suitable for disjunct populations of arctic-alpine plants since deglaciation. While lake-mediated cooling is a well-known phenomenon, the degree and extent of temperature buffering at most lakes is unknown. In this thesis, I quantify temperature buffering for a network of lakes across the boreal forest region of western Canada, identify environmental factors affecting buffering, and assess the effect on cold-adapted plant species. Using Lake Superior as a model system, I first sought to explain the spatial patterns and environmental drivers of disjunct arctic-alpine plant refugia on its shores under current conditions and projected future warming scenarios, using known occurrence records as indicators of refugia. I used an ecological niche model (ENM) to predict the presence of all disjunct species and individual species distribution models (SDMs) for species-specific predictions for 20 common disjunct species to identify hotspots of disjunct species richness. I found that bedrock type, elevation above the water, distance from shore, mean July daytime temperature, and the depth of near-shore waters best predicted disjunct occurrence, and that the highly exposed north-central and north-east shore of the lake had the highest richness of disjunct species. I predicted 2236 km of the shoreline (51%) as potentially suitable disjunct refugia habitat today, but this was reduced to 20% and 7% with moderate (894 km) and warmest (313 km) climate warming scenarios. Second, I determined the effects of macroclimatic variables and localized site conditions on disjunct arctic-alpine plant species richness. In the field, I measured macroclimate variables and local site conditions for 43 shoreline sites across Lake Superior’s north shore. I fitted a piecewise structural equation model (SEM) and found that disjunct richness was directly affected by shore width, elevation above the water, mean July temperature, and bedrock type. I found that richness of common boreal species was not significantly correlated with richness of disjunct arctic-alpine species, and thus that current open bedrock habitats are not limited by competition. Third, I examined the extent, degree, and patterns of temperature buffering at Lake Superior using temperature logger transects away from shorelines, and predicted drivers of buffering using generalized linear models (GLMs). I found the magnitude of cooling varied geographically, with the exposed north-central shore showing the highest temperature anomaly from interior reference sites. In this region, shoreline sites were 5.8 °C cooler than interior reference sites during July, and had a shortened growing season. The nearest similar July temperatures are found approximately 1000 km further north (~54°N). Temperature anomaly was best predicted by elevation above the water, shore width, nearshore July water surface temperature, and nearshore water depth. Finally, I examined the temperature buffering effects at a network of 11 lakes across western Canada’s boreal forest. Using the same study design as at Lake Superior, I predicted the effects of varying lake characteristics on mean July temperature anomaly, spring ice-off period and fall ice-on period temperature anomaly, and the proportion of coniferous trees at sites. July temperatures were coolest on the downwind side of lakes, within 10 km of shore, and at high volume lakes. Ice-off temperature anomalies were best predicted by the interaction between lake area and lake depth, while the proportion of coniferous trees at sites was best predicted by average July temperature. Knowledge of the extent and magnitude of lake-mediated cooling and drivers of lakeshore refugia at lakes across western boreal Canada may inform regional climate models, as well as guide for conservation and management of cold-adapted species, and protected areas under a changing climate.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.