| When was the current long-range forecast system implemented? | November 2017 |
|---|---|
| Is it a coupled forecast system? | Yes |
| Is it a Tier-2 forecast system? | No |
| Atmospheric model resolution | Spectral, Tco319/L91 (horizontal resol. about 35Km) |
| Ocean model (if applicable) | NEMO version3.4 |
| Source of atmospheric initial conditions | ERA-interim for the period 1981 to 2016 and ECMWF operational analysis from 2017 onwards |
| Source of ocean initial conditions | Global Ocean Data Assimilation System (NEMOVAR -OCEAN5) |
| If Tier-2, what is the source of SST predictions? | N/A |
| Hindcast period | 1981-2016 however the calibration period is: 1993-2016 |
| Ensemble size for the hindcasts | 25 |
| How is the hindcast ensemble configured? | 5-member ensemble ocean analysis combined with SST perturbations and the activation of stochastic physics |
| Ensemble size for the forecast | 51 |
| How is the forecast ensemble configured? | 5-member ensemble ocean analysis combined with SST perturbations and the activation of stochastic physics |
| Length of forecasts | 7 months (four times a year 11 members are extended to 13 months) |
| Data format | GRIB 1 |
| What is the latest date predicted anomalies for the next month/season become available? | 10 th of the current month |
| How are the forecast anomalies constructed? | Departures from the model climate estimated by the hindcast integrations (1993-2016) |
| URL where forecast are displayed | https://www.ecmwf.int/en/forecasts/documentation-and-support/long-range |
| Point ofContact | Laura Ferranti (l.ferranti@ecmwf.int) |
