Debating how to assess hydrological model uncertainty and weaknesses



The projections of hydrological models, as numerical abstractions of the complex systems they seek to represent, suffer from epistemic uncertainty due to approximation errors in the model, incomplete knowledge of the system, and, in more extreme cases, flawed underlying theories or faulty data. These errors can result in complex nonstationary biases in model predictions. Improving hydrological models—whether they be dynamic models attempting to represent the physical system from first principles or statistical, data-driven models—depends on having a way to determine the amount of uncertainty associated with the model's projections and a reliable way to deduce which model components or forcing data are responsible for it.