Here, we explored in depth the relationship among the deterministic prediction skill, the probabilistic prediction skill and the potential predictability. This was achieved by theoretical analyses and, in particular, by an analysis of long-term ensemble ENSO hindcast over 161 years from 1856 to 2016. First, a nonlinear monotonic relationship between the deterministic prediction skill and the probabilistic prediction skill, derived by theoretical analysis, was examined and validated using the ensemble hindcast. Further, the co-variability between the potential predictability and the deterministic prediction skill was explored in both perfect model assumption and actual model scenario. On these bases, we investigated the relationship between the potential predictability and probabilistic prediction skill from both the practice of ENSO forecast and theoretical perspective. The results of the study indicate that there are nonlinear monotonic relationships among these three kinds of measures. The potential predictability is considered to be a good indicator for the actual prediction skill in terms of both the deterministic measures and the probabilistic framework. The relationships identified here exhibit considerable significant practical sense to conduct predictability researches, which provide an inexpensive and moderate approach for inquiring prediction uncertainties without the requirement of costly ensemble experiments.

Temporal vertical eddy viscosity coefficient (VEVC) in an Ekman layer model is estimated using an adjoint method. Twin experiments are carried out to investigate the influences of several factors on inversion results, and the conclusions of twin experiments are 1) the adjoint method is a capable method to estimate different kinds of temporal distributions of VEVCs; 2) the gradient descent algorithm is better than CONMIN and L-BFGS for the present problem, although the posterior two algorithms perform better on convergence efficiency; 3) inversion results are sensitive to initial guesses; 4) the model is applicable to different wind conditions; 5) the inversion result with thick boundary layer depth (BLD) is slightly better than thin BLD; 6) inversion results are more sensitive to observations in upper layers than those in lower layers; 7) inversion results are still acceptable when data noise exists, indicating the method can sustain noise to a certain degree; 8) a regularization method is proved to be useful to improve the results for present problem; and 9) the present method can tolerate the existence of balance errors due to the imperfection of governing equations. The methodology is further validated in practical experiments where Ekman currents are derived from Bermuda Testbed Mooring data and assimilated. Modeled Ekman currents coincide well with observed ones, especially for upper layers. The results demonstrate that the assumptions of depth dependence and time dependence are equally important for VEVCs. The feasibility of the typical Ekman model, the imperfection of Ekman balance equations, and the deficiencies of the present method are discussed. This method provides a potential way to realize the time variations of VEVCs in ocean models.

No abstract