Uncertainty Transformation in Ecological Simulation Models

Being simplified representations of reality, simulation models can never be perfect and their results are always somewhat uncertain. That is why quantification of model uncertainty is important during interpretation of simulation results in decision making process. Uncertainty describes deviations of simulated ecosystem’s characteristics from known or observed values. Several sources contribute to such deviations including those associated with main model components such as forcing functions, mathematical formulae, parameters and universal constants and intrinsic model features. Model uncertainty can be evaluated based on its linear estimate under the assumption that all sources of uncertainty are independent. Traditional approaches to investigating model uncertainty consider individual sources whose contribution to the uncertainty can be quantified for a given task. Although the result is incomplete it helps to improve the understanding of the model and increase the confidence in simulation results.