Modeling boreal forest response to climate variability in central Canada

This study examines the importance of short-term climate variability when simulating forest succession using ecological process models. A version of the FORSKA2 forest gap model was modified for use with daily climate data and applied along a transect of sites crossing the boreal region in central Canada, including the aspen-parkland and forest-tundra ecotones where impacts of climatic change on forest ecosystems could be particularly significant. The model's sensitivity to forcing with daily climate observations compared to monthly mean and long term averages of monthly mean climate data was investigated. Inclusion of daily climate (minimum and maximum temperature and total precipitation) improved the simulations of key characteristics of present-day forest along the transect, and was particularly important at the ecotones. The results demonstrate that changes in variability associated with future change in mean climate are likely to be important when trying to predict boreal forest responses to projected future climate change. Ideally, the use of projected daily climate data or data based on the statistical characteristics of daily climate is highly recommended for future impact studies. A number of approaches to further improve the functioning of the model are also presented.