We used the LANDIS-II forest landscape disturbance and succession model with an extension including carbon and nitrogen cycling to assess the immediate future of carbon cycling in northern Michigan, and to test the effects of climate change and land management scenarios on future carbon fluxes. We constrained the model using data resources at the University of Michigan Biological Station (UMBS), including eddy-covariance flux measurements and a forest chronosequence created from experimental clear-cut harvesting and burning. Climate change effects were tested by applying downscaled climate predictions as the model's climate forcing, and comparing the results to model runs that use present average climate.
We used the UMBS data resources to constrain and evaluate model performance for the forest types represented in the UMBS area, and then ran the model into the future to forecast evolution of the northern Michigan forest carbon sink. Model results showed that forest carbon uptake will decline in the near future as the first generation of early successional trees dies. The forest will briefly be a source of carbon to the atmosphere, but carbon uptake rates will recover as the next generation of later successional trees grows. However, the shift of the forest to a mixed age, later successional character means that the total carbon sink will not recover to present levels. Climate change simulations suggest that both ecosystem respiration and productivity will increase, resulting in a small net change in carbon uptake.