Find out about us: our forecasts

Medium-Range High-Resolution Forecast	Medium-Range Ensemble Forecast	Long-Range Forecast

Operational forecasts

ECMWF produces a suite of operational forecasts for various lead times:

• Medium-range forecast: comprises the high-resolution and the ensemble forecasts of weather, at the space and time-scales represented by the relevant model, up to 10 and 15 days ahead, respectively, and the associated uncertainty.
• Extended-range (monthly) forecast: comprises ensembles of individual forecasts and post-processed products of average conditions (e.g. weekly averages) up to 1 month ahead, and the associated uncertainty.
• Long-range (SEAS) forecast: comprises ensembles of individual forecasts and post-processed products of average conditions (e.g. monthly averages) up to 13 months ahead, and the associated uncertainty.

In addition re-forecasts are calculated operationally using the current system configuration but applied to the weather over past decades:

• Re-forecasts: comprise forecasts run for past decades necessary to estimate the model climate and the level of skill and to generate some of the operational products.

These forecasts are produced using the ECMWF Integrated Forecasting System (IFS).

Component models

There are five component models of the IFS.

• Atmospheric model with various configurations suited to the space scale and time range of the required forecasts. The current configurations for High Resolution (HRES), Ensemble (ENS) and Boundary Conditions (BC) are given in Table 1a.
• Ocean wave model is a version of the WAve Model (WAM) model which has been further developed in house. It is coupled to the atmospheric model or run as a standalone model in the Limited-Area Wave (LAW) configuration.
• Ocean model is a version of the NEMO (Nucleus for European Modelling of the Ocean) model.
• Process models are used to describe, for example, land-surface processes, surface ocean waves, and sea-ice.
• Perturbation models are used to simulate the effect of uncertainties in the observations, initial conditions, surface boundary conditions, and modelled processes. These produce perturbations for use in ensemble forecasts.

Component analyses

In addition, there are five components for analysing the state of the atmosphere and oceans.

• 4DVAR (4-Dimensional Variational analysis) provides a detailed estimate of the current state of the atmosphere computed utilising as optimally as possible observations and prior information about the Earth-system using ECMWF’s highest resolution model.
• EDA (Ensemble of Data Assimilations) provides an ensemble of estimates of the current state of the atmosphere and its uncertainty. The EDA estimate of the analysis uncertainty can be used as an approximation of the 4DVAR uncertainty. The current configuration of EDA is given in Table 1a.
• ORTA (Ocean Real-Time Analysis) provides an estimate of the ocean initial state and its uncertainty. The current configuration of ORTA is given in Table 1c.
• ERA (ECMWF Reanalysis) provides consistent estimates of the state of the atmosphere generated using a fixed, lower-resolution version of 4DVAR for the past decades. The latest ERA product, ERA-Interim, covers the period since 1979 and is continued in real time to support climate monitoring. ERA-Interim is also used to define the atmospheric initial conditions of the re-forecasts.
• ORA (Ocean Reanalysis) is the equivalent of ERA for the oceans.

Key characteristics of the forecasting system