A process-based model to simulate growth in forests with complex structure: Evaluation and use of 3D-CMCC Forest Ecosystem Model in a deciduous forest in Central Italy

Forest ecosystems are characterized by high spatial heterogeneity, often related to complex composition and vertical structure which is a challenge in many process-based models. The need to expand process-based models (PBMs) to take in account such structural complexity led to development and testing of a new approach into Forest Ecosystem Models (FEMs), named 3D-CMCC-FEM, able to investigate carbon and water fluxes, including biomass pools and their partitioning, for complex multi-layer forests. 3D-CMCC FEM integrates several characteristics of the functional–structural tree models and the robustness of the light use efficiency (LUE) approach to investigate forest growth patterns and yield processes. The modelling approach was tested by simulating the effects of competition for light and water, growth and yield of a two-layered deciduous forest dominated by Turkey Oak in central Italy for a period of eight years. The model outputs were validated against a series of independently measured data for the major biomass pools, the inter-annual stem increments and above-ground net primary productivity of the overstorey and understorey, respectively. The comparison of Leaf Area Index, Gross Primary Production, and evapotranspiration produced by the model against MODIS data showed agreement in results. In addition, the multi-layered model approach was evaluated against a series of simplified versions to determine whether the enhanced complexity of the model positively contributed to its predictive ability. The proposed model reduced the error in the estimates of forest productivity (e.g. NPP) and dynamics (e.g. growth, mortality) and indicates the importance of considering, as far as possible, the structural complexity in PBMs.