Predicting climate change effects on vegetation, soil thermal dynamics, and carbon cycling in ecosystems of interior Alaska

A dynamic vegetation model has been used to predict patterns of recent past and potential future change in taiga forest ecosystems of interior Alaska. The model, called CASA (Carnegie Ames Stanford Approach), is a process-based ecosystem depiction of plant and soil processes, including all major cycles of water and carbon. CASA’s dynamic vegetation component is intended to facilitate coupling to general circulation models of the atmosphere, which require mechanistic fluxes and feedbacks from terrestrial vegetation. Simulation results for selected Alaska sites of Denali National Park suggest that the past 50-year climate trends of warming temperatures may shift the taiga ecosystem from dominance by evergreen needleleaf trees to a more mixed broadleaf–needleleaf tree composition. For other (higher elevation) areas of Denali, our model predicts that a difference of only about 3 °C in mean annual air temperatures appears to differentiate the permanent presence of tundra vegetation forms over taiga forest. The model predicts that over the 1950–1999 climate record at Denali station, the changing taiga ecosystems were net sinks for atmospheric CO2 of about 1.3 kg C m−2. During the warm 1990s, these forests were predicted to be net carbon sinks of more than 15 g C m−2 per year in 8 out of 10 years. Predicted NPP for the forest continues to increase with a projected warming trend for the next 25 years at a mean rate of about +1.2 g C m−2 per year. On the basis of these model results, a series of crucial field site measurements can be identified for inclusion in subsequent long-term ecological studies of the changing taiga forest.