11B.2
A consolidated CLIPER model for improved August-September ENSO prediction skill
Benjamin Lloyd-Hughes, University College London, Dorking, Surrey, United Kingdom; and M. A. Saunders and P. Rockett
A prime challenge for ENSO seasonal forecast models is to predict boreal summer ENSO conditions at lead. August-September ENSO has a strong influence on Atlantic hurricane activity, Northwest Pacific typhoon activity and tropical precipitation. However, summer ENSO skill is low due to the spring predictability barrier during March-May. The statistical ENSO-CLIPER (CLImatology and PERsistence) prediction model is arguably one of the more successful ENSO seasonal forecast models to date. The sensitivities of the CLIPER model to teleconnected predictor averaging period (1, 2, 3, 4, 5 and 6 months; standard CLIPER uses only 3 months) and to the variance factor used during the optimal combination of predictors (1%, 2.5% and 5% improvement factors; standard CLIPER uses only the 2.5% factor) are examined. A 'Consolidated' ENSO-CLIPER model is defined as the mean of an ensemble of 18 models built using these 6 averaging periods and 3 improvement factors. Comparing the August-September 1952-2002 cross-validated hindcast skill from the consolidated and standard CLIPER models shows that the consolidated model outperforms the standard model by 10-20% in absolute percentage mean square error improvement over climatology at all leads from 2 to 6 months for all the main ENSO indices (3, 3.4 and 4). The consolidated CLIPER August-September 51-year hindcast skill is also positive to 97.5% confidence at leads out to 4 months (early April) for all ENSO indices. Optimisation of the consolidated CLIPER model may lead to further skill improvements. .
Session 11B, interannual variablity II (Note new session start time from the original published program)
Wednesday, 5 May 2004, 10:15 AM-11:30 AM, Napoleon I Room
Previous paper Next paper