The ECMWF ensemble prediction system

There are two key sources of forecast error: the presence of uncertainties in the initial conditions and the approximate simulation of atmospheric processes achieved in the state-of-the-art numerical models. These two sources of uncertainties limit the skill of single, deterministic forecasts in an unpredictable way, with days of high/poor quality forecasts randomly followed by days of high/poor quality forecasts. One way to overcome this problem is to move from a deterministic to a probabilistic approach to numerical weather prediction, and try to estimate the time evolution of an appropriate probability density function in the atmosphere’s phase space. Ensemble prediction is a feasible method to estimate the probability distribution function of forecast states. The European Centre for Medium-Range Weather Forecasts (ECMWF) Ensemble Prediction System (EPS) is one of the most successful global ensemble prediction systems run on a daily basis. In this chapter the ECMWFEPS is described, its forecast skill documented, and potential areas of future development are discussed. The rationale behind ensemble prediction The time evolution of the atmospheric flow, which is described by the spatial distribution of wind, temperature, and other weather variables such as specific humidity and surface pressure, can be estimated by numerically integrating the mathematical differential equations that describe the system time evolution. These equations include Newton’s laws of motion used in the form ‘acceleration equals force divided by mass’ and the laws of thermodynamics. Numerical time-integration is performed by replacing time-derivatives with finite differences, and spatial-integration either by finite difference schemes or spectral methods.