Simulated long-term effects of harvest and biomass residue removal on soil carbon and nitrogen content and productivity for two Upper Great Lakes forest ecosystems

We used the ecosystem process model Biome-BGC to simulate the effects of harvest and residue removal management scenarios on soil carbon (C), available soil nitrogen (N), net primary production (NPP), and net ecosystem production (NEP) in jack pine (Pinus banksiana Lamb.) and sugar maple (Acer saccharum Marsh) ecosystems in northern Wisconsin, USA. To assess harvest effects, we simulated short (50-year) and long (100-year) harvest intervals, high (clear-cut) and low (selective) harvest intensities, and three levels of residue retention (15%, 25%, and 35%) over a 500-year period. The model simulation of NPP, soil C accumulation, and NEP agreed reasonably well with biometric and eddy-covariance measurements of these two ecosystems. The more intensive (50-year rotation clear-cuts with low residue retention) harvest scenarios tended to have the greatest NEP (420 and 678 t C ha-1 for the 500-year interval for jack pine and sugar maple, respectively). All the harvest scenarios decreased mineral soil C and available mineral soil N content relative to the no-harvest scenario for jack pine and sugar maple. The rate of change in mineral soil C decreased the greatest in the most intensive biomass removal scenarios (-0.012 and -0.072 t C ha-1 yr-1 relative to no-harvest for jack pine and sugar maple, respectively) and the smallest decrease was observed in the least intensive biomass removal scenarios (-0.002 and -0.009 t C ha-1 yr-1 relative to no-harvest for jack pine and sugar maple, respectively). The more intensive biomass removal harvest scenarios in sugar maple significantly decreased peak productivity (NPP) in the simulation period.