Using a forest patch model to predict the dynamics of stand structure in Swiss mountain forests

Forest patch models have been applied to simulate forest development and long-term forest succession in many studies. The main focus of these simulations has been on species composition and biomass in natural forests, but these models could also become useful for the prediction of other structural forest patterns such as size distributions. Up to now. most of these models have been validated by approaches such as comparison of simulation results with potential natural vegetation (PNV), or national forest inventory data. While these approaches may be appropriate to validate the simulated species composition, they are not sufficient in testing the prediction of other structural patterns. Thus, little is known about the accuracy of forest patch models in simulating structural forest patterns such as size distribution of different forest types. For this reason. we tested the forest patch model ForClim against empirical data from three Swiss mountain forests. The objectives of this study were: (i) to investigate the performance of ForClim in simulating structural forest patterns and (ii) to assess the influence of the regeneration. growth and mortality submodels of ForClim on the simulation results. Several shortcomings of the model were identified and quantified. In particular. the stress-induced mortality, implemented in ForClim was found to overestimate the actual mortality rates. The excessive mortality was most likely caused by an inaccurate growth function or an overestimation of light competition. Once the stress-induced mortality was reduced, ForClim was able to reproduce structural forest patterns in an accurate manner. Based on these encouraging results, we suggest that ForClim as well as other forest patch models could become important tools for further applications in forest research. (C) 2004 Elsevier B.V. All rights reserved.