W.J. de Groot, P.M. Bothwell, D. Carlsson, K. Logan, Ross W. Wein, C. Li
The effects of future fire regimes altered by climate change, and fire management in adaptation to climate change were studied in the boreal forest region of the Prairie provinces. Four National Parks were used asstudy areas. Present (1975-90) and future (2080-2100) fire regimes were simulated in Wood Buffalo National Park, Elk Island National Park, Prince Albert National Park and Riding Mountain National Park using data from the Canadian (CGCM1) and Hadley (HadCM3) Global Climate Models (GCM) in separate simulation scenarios. The long-term effects of the different fire regimes on forests were simulated using astand-level, boreal fire effects model (BORFIRE) developed for this study. Changes in forest composition and biomass storage due to future altered fire regimes were determined by comparing current and future simulation results. This was used to assess the ecological impact of altered fire regimes on boreal forests,and the future role of these forests as carbon sinks or sources. Additional future simulations were run using adapted fire management strategies to meet the management goals for each National Park. This included increased fire suppression and the use of prescribed fire to meet fire cycle objectives. Future forest composition and biomass storage under current and adapted fire management strategies were also compared to determine the impact of various future fire management options.
Both of the GCM’s showed more severe burning conditions under future fire regimes. This includes fires with higher intensity, greater depth of burn, greater total fuel consumption and shorter fire cycles (or higher rates of annual area burned). The Canadian GCM indicated burning conditions more severe than the Hadley GCM. The Canadian GCM results also appeared more reliable when fire weather output was compared to current and historical data.
In the model simulations, the shorter fire cycles of future fire regimes generally favoured aspen and birch because of their post-fire resprouting ability, and jack pine because of its serotinous cones which release stored seed after fire. Shorter fire cycles provided more frequent regeneration opportunity for these species. Because black spruce is an annual seeder and has semi-serotinous cones, regeneration was only minimally influenced by future changes in the fire cycle. However, white spruce stands declined sharply due to shorter fire cycles, and the spring and summer fire regime of the study areas. This was because white spruce doesn’t store seed and seed ripening doesn’t occur until late summer or early fall, so there is no opportunityto regenerate when trees are killed by early season fire. For all of the study area Parks, maintaining representation of pure and mixed white spruce ecosystems will be a concern under future fire regimes. The model simulations also showed that a management goal of fire exclusion would effectively lead to the removal of jack pine from the study areas, and cause a sharp decline in aspen and birch stands.
There was a general increase in total biomass storage under the simulated future fire regimes. This was caused by two factors. Shorter fire cycles resulted in a younger age-class distribution so there were less slow-growing, low density old stands, and more fast-growing, high density young stands. The second factor was an increase in aspen, which is faster growing than the other species. Aspen regeneration was favoured by short fire cycles, and aspen seedlings out-competed jack pine and white birch. As well, when white spruce failed to regenerate, mixed stands converted to pure aspen stands. A secondary effect of greater aspen live tree biomass was an increase in forest floor biomass because of increased detrital input. Biomass storage was very low in the fire exclusion simulations.
In Wood Buffalo National Park, simulations of increased future fire suppression assisted in maintaining white spruce ecosystems and older age classes of all species, but it had a minimal impact on representation of other forested ecosystems. Increased fire suppression also increased the long-term total biomass storage in the Park by 83M tonnes.
The simulations showed prescribed burning to be an important component of future fire management in Elk Island National Park, Prince Albert National Park and Riding Mountain National Park. Without prescribed fire, the fire cycle in all three Parks would be too long to maintain current stands of aspen, jack pine and white birch. A range of fire regimes appears necessary to manage different areas in each National Park. Aspen in open or closed stands can be promoted by burning after vernal leaf flush, or its removal can be facilitated by burning prior to leaf flush with short to moderate fire cycles or 25-75 years. The use of short fire cycles may be required to maintain grasslands and shrublands and prevent the encroachment of aspen.
White spruce stands require fires of low intensity, or late season fires and longer fire cycles (100+ years). In Prince Albert National Park and Riding Mountain National Park , the use of prescribed fire on 75-year and 100-year fire cycles to maintain current forest ecosystems resulted in a total biomass storage increase of 10-17M tonnes and 8-12M tonnes, respectively.
Future needs in fire and climate change research includes development of landscape models that simulate physical and ecological fire effects, and analysis of future fire management strategies in the commercial forest zone.