Coastal catchments in British Columbia, Canada, experience a complex mixture of rainfall- and snowmelt-driven contributions to flood events. Few operational flood-forecast models are available in the region. Here, we integrated a number of proven technologies in a novel way to produce a super-ensemble forecast system for the Englishman River, a flood-prone stream on Vancouver Island. This three-day-ahead modeling system utilizes up to 42 numerical weather prediction model outputs from the North American Ensemble Forecast System, combined with six artificial neural network-based streamflow models representing various slightly different system conceptualizations, all of which were trained exclusively on historical high-flow data. As such, the system combines relatively low model development times and costs with the generation of fully probabilistic forecasts reflecting uncertainty in the simulation of both atmospheric and terrestrial hydrologic dynamics. Results from operational testing by British Columbia's flood forecasting agency during the 2013-2014 storm season suggest that the prediction system is operationally useful and robust.

This article has been retracted; see doi:10.1002/2014WR015352

•We develop a short-term ensemble reservoir inflow forecasting system.•The ensemble attempts to sample all sources of uncertainty in the modeling chain.•Increasing the diversity of the ensemble greatly improves ensemble quality.•Bias correction of ensemble members offers significant additional improvement.•For the flashy case study basin, only recent errors are important in bias correction.

This article examines the current practice of streamflow modelling, a field under development for over a century. A sample of the wide range of assessment and planning applications of streamflow models is presented. The diversity in the use of these models is mirrored in the diversity of model complexity, and modelling approaches ranging from empirical to physically based and from lumped to fully distributed are described with examples. Predictions derived from hydrological models are subject to many sources of error; these are discussed along with methods for error minimization or anticipation. Model error is generally quantified using an ensemble of forecasts meant to sample the range of predictive uncertainty. This ensemble can be used to generate reliable probabilistic forecasts of hydrological quantities if all sources of error are accounted for. To date, applications of ensemble methods in streamflow forecasting have typically focused on only one or two error sources. A challenge will be to develop ensemble streamflow forecasts that sample a wider range of predictive uncertainty.