Characterizing Climate Model Uncertainty Using an Informal Bayesian Framework: Application to the River Nile

Abstract

Assessing climate change effects on water resources is the first step in preparing climate change adaptation measures. However, this is often clouded by the large range of uncertainty resulting from a long chain of modeling activities. Despite progress made to improve climate models, downscaling methods, and hydrological models, uncertainties will remain. This paper proposes a framework to propagate and quantify the uncertainty from the different sources that can be applied at the full cascade but focuses on the climate-modeling component, i.e., different climate models and emissions scenarios. This framework is based on the generalized likelihood uncertainty estimation (GLUE) methodology, which is widely used in the hydrologic community but has not been applied as such to climate impact modeling. This paper presents a preliminary application of the proposed framework to the flow of the main Nile at Dongola.