On the Evaluation of Climate Change Impact Models for Adaptation Decisions

Abstract

AbstractDetailed understanding of the potential local or regional implications of climate change is required to guide decision- and policy-makers when developing adaptation strategies and designing infrastructure solutions suitable for potential future conditions. Impact models that translate potential future climate conditions into variables of interest (such as drought or flood risk) are needed to create the required causal connection between climate and impact for scenario-based analyses. Recent studies suggest that the main strategy for the validation of such models (and hence the justification for their use) still heavily relies on the comparison with historical observations. In this short paper, the author suggests that such a comparison alone is insufficient and that global sensitivity analysis provides additional possibilities for model evaluation to ensure greater transparency and better robustness of model-based analyses. Global sensitivity analysis can be used to demonstrate that the parameters defining intervention options (such as land use choices) adequately control the model output (even under potential future conditions); it can be used to understand the robustness of model outputs to input uncertainties over different projection horizons, the relevance of model assumptions, and how modelled environmental processes change with climatic boundary conditions. Such additional model evaluation would strengthen the stakeholder confidence in model projections and therefore into the adaptation strategies derived with the help of these model outputs.