ALBERT

Description: An information theory-based framework is developed to assess the predictability and quantify the forecast uncertainty of ENSO complexity, which includes different types of ENSO events with diverse characteristics. With the assistance of a recently developed multiscale stochastic conceptual model that successfully captures both the large-scale dynamics and many crucial statistical properties of the observed ENSO complexity, it is shown that different ENSO events possess distinct predictability limits. Beyond the ensemble mean value, the spread of the ensemble members also contains valuable information about predictability. First, La Niña events are most predictable at long lead times, especially as a subsequent transition after eastern Pacific (EP) El Niño events or during multi-year La Niña phases. Second, EP El Niños tend to be more predictable than the central Pacific (CP) El Niño events up to one year ahead due to a more favorable signal-to-noise ratio, even though their onset remains hard to predict. Third, 4 out of 6 CP El Niño events seem to be predictable up to 24 months ahead, where such strong predictability is often converted to skillful forecast. Fourth, strengthening/weakening the Walker circulation intensity increases/decreases CP predictability at long leads. Fifth, accounting for intraseasonal wind events in the initial condition strongly contributes to EP predictability at lead times of less than one year. Finally, it is shown that a Gaussian approximation of the information gain computation is accurate, making the information theory approach tractable for studying the predictability of more sophisticated models.

Description: The discrete element method (DEM) provides a new modeling approach for describing sea ice dynamics. It exploits particle-based methods to characterize the physical quantities of each sea ice floe along its trajectory under Lagrangian coordinates. One major challenge in applying DEM models is the heavy computational cost when the number of floes becomes large. In this paper, an efficient Lagrangian parameterization algorithm is developed, which aims at reducing the computational cost of simulating the DEM models while preserving the key features of the sea ice. The new parameterization takes advantage of a small number of artificial ice floes, called the superfloes, to effectively approximate a considerable number of the floes, where the parameterization scheme satisfies several important physics constraints. The physics constraints guarantee the superfloe parameterized system will have short-term dynamical behavior similar to that of the full system. These constraints also allow the superfloe parameterized system to accurately quantify the long-range uncertainty, especially the non-Gaussian statistical features, of the full system. In addition, the superfloe parameterization facilitates a systematic noise inflation strategy that significantly advances an ensemble-based data assimilation algorithm for recovering the unobserved ocean field underneath the sea ice. Such a new noise inflation method avoids ad hoc tunings as in many traditional algorithms and is computationally extremely efficient. Numerical experiments based on an idealized DEM model with multiscale features illustrate the success of the superfloe parameterization in quantifying the uncertainty and assimilating both the sea ice and the associated ocean field.