Hydrological Forecasting

Hydrological models underpin many hydrometeorological forecasting techniques and include rainfall-runoff, flow routing and statistical approaches. Conceptual rainfall-runoff models are the most widely used in real-time forecasting, but physically based and data-driven models have an important role too, along with flow routing techniques encompassing both hydrological and hydrodynamic approaches. Ensemble and probabilistic techniques are increasingly used to estimate the uncertainty in outputs. Additional modelling components may be required to represent specific features of a catchment such as reservoirs, lakes and wetlands. For real-time use, models are typically operated within an automated forecasting system which controls the gathering of data, scheduling of model runs and data assimilation. This chapter provides an introduction to these various techniques along with operational considerations for real-time use, including data validation, pre- and post processing of outputs, and the important topics of forecast uncertainty and forecast verification.