Effects of error in model input: experiments with a forest patch model

Dynamic vegetation models are increasingly applied in spatially explicit impact assessments. Both, in model evaluation studies as well as in model application it is important to acknowledge the uncertainty in simulation results. In this study we analyse how error in estimates of soil and climate data required to initialise and drive the patch model PICUS v1.2 propagates through the model and affect simulated species composition and biomass. From the results it can be concluded that (1) the effect varies greatly due to ecological characteristics of a particular site and to involved tree species, and (2) uncertainty in model input, if considered appropriately, might substantially hamper the chance to falsify model assumptions. For spatially explicit modelling studies it is recommended to place more emphasis on the effect of uncertainties in soil and climate data used to initialise and drive ecosystem models.