New approach for optimal perturbation method in ensemble climate prediction with empirical singular vector Jong-Seong Kug, Yoo-Geun Ham, Masahide Kimoto, Fei-Fei Jin, In-Sik Kang Abstract: In this study, a new method is developed to generate optimal perturbations in ensemble climate prediction. In this method, the optimal perturbation in initial conditions is the 1st leading singular vector, calculated from an empirical linear operator based on a historical model integration. To verify this concept, this method is applied to a hybrid coupled model. It is demonstrated that the 1st leading singular vector from the empirical linear operator, to a large extent, represents the fast-growing mode in the nonlinear integration. Therefore, the forecast skill with the optimal perturbations is improved over most lead times and regions. In particular, the improvement of the forecast skill is significant where the signal-to-noise ratio is small, indicating that the optimal perturbation method is effective when the initial uncertainty is large. Therefore, the new optimal perturbation method has the potential to improve current seasonal prediction with state-of-the-art coupled GCMs. PDF: P2010_5