Calibrating a regional growth and yield model for statewide forest carbon assessment in four common forest types in Minnesota, USA

Understanding how forest carbon is stored and sequestered requires a quantitative understanding of how different forest management strategies shape the structure and composition of forests. Essential to this includes parameterizing growth and yield models to depict complex silvicultural strategies and calibrating them to localized stand conditions and growth observations. This project used Forest Inventory and Analysis (FIA) data as inputs to the Lake States variant of the Forest Vegetation Simulator (FVS-LS) to calibrate and forecast forest carbon attributes in Minnesota, USA for 100 years across the four most commonly managed forest types in the state (aspen/birch, red pine, northern hardwoods, and lowland conifers) and four forest management scenarios: (1) no management, (2) business as usual, (3), economic intensive, and (4) climate-adapted. Previous measurements collected from trees on FIA plots were used to calibrate the mortality and growth modifiers and add an appropriate number of regenerating trees to the FVS-LS simulations to closely approximate forest development observed in Minnesota. Simulations showed that forests in the no management and climate-adapted scenarios generally contained the greatest carbon stocks (including above and belowground carbon in the forest and in harvested wood products) at the end of the 100-year simulation. Across the 100-year simulation for all four forest types, average carbon stock change was highest in the climate-adapted and no management scenarios (1.36 and 1.26 Mg CO2-eq ha−1 yr−1, respectively) followed by business as usual and economic intensive scenarios (1.20 and 0.78 Mg CO2-eq ha−1 yr−1, respectively).These results show the opportunity for calibrating existing growth and yield models for forest carbon assessments and how different forest management scenarios can lead to divergent forest carbon outcomes when simulated on a statewide basis.