Taking account of solutions from ECMWF and other models - Grand Ensembles

An ensemble of NWP model runs using different models but starting from the same data time is known as a grand ensemble.  Combining the current ensembles from different models into a larger ensemble (as in Fig6.1.1.1-1) can show consistency or divergence in forecasts.  They can also show alternative solutions.  Nevertheless, there can be some agreement between the spread and jumpiness. 

One shouldn't rely on the results of any individual ensemble member or ensemble run itself.  One shouldn't rely on the results of any individual ensemble member or ensemble run alone.  If there is any indication of extreme or hazardous weather in the ensembles then that threat should be passed to users, but with a very low probability.  


Fig6.1.1.1-1: Schematic illustration of the relation between:

  • the latest Ensemble forecasts (green lines).  Broadly in agreement with latest Ensemble Control Forecast (ex-HRES).
  • the two latest Ensemble Control Forecast (ex-HRES) solutions (red lines).  Broadly in agreement with latest ensemble solution.
  • results from other forecast models (blue lines).  These show some more extreme troughs or higher maxima.

The latest Ensemble Control Forecast (ex-HRES) and ensemble solutions are rather more likely.  However, any solutions from other model runs that show greater variability or are more extreme should not be ignored.  These results should be retained as a small probability.