Using ensembles for short-range forecasting

Numerical forecasts from a pilot program on short-range ensemble forecastin g at the National Centers for Environmental Prediction are examined. The en semble consists of 10 forecasts made using the 80-km Era Model and 5 foreca sts from the regional spectral model. Results indicate that the accuracy of the ensemble mean is comparable to that from the 29-km Meso Eta Model for both mandatory level data and the 36-h forecast cyclone position. Calculati ons of spread indicate that at 36 and 48 h the spread from initial conditio ns created using the breeding of growing modes technique is larger than the spread from initial conditions created using different analyses. However, the accuracy of the forecast cyclone position from these two initialization techniques is nearly identical. Results further indicate that using two di fferent numerical models assists in increasing the ensemble spread signific antly. There is little correlation between the spread in the ensemble members and the accuracy of the ensemble mean for the prediction of cyclone location. S ince information on forecast uncertainty is needed in many applications, an d is one of the reasons to use an ensemble approach, the lack of a correlat ion between spread and forecast uncertainty presents a challenge to the pro duction of short-range ensemble forecasts. Even though the ensemble dispersion is not found to be an indication of for ecast uncertainty, significant spread can occur within the forecasts over a relatively short time period. Examples are shown to illustrate how small u ncertainties in the model initial conditions can lead to large differences in numerical forecasts from an identical numerical model.