| Summary: | Satellite and in-situ observations are now routinely combined with numerical models in order to estimate the time-evolving oceanic circulation and to address a wide variety of operational and research problems. For climate dynamics analysis, what is required is a synthesis of all avail-able observations over the last several decades with the best possible numerical model. Rigorous low-resolution estimates of ocean circulation are already possible using the existing data base and modeling capability. But these low-resolution estimates lack the ability to resolve many small-scale oceanic processes, for example, ow over narrow sills, western boundary currents, regions of deep convection, and eddies, that are important both for climate studies and for operational applications. I will discuss four recent advances that bring rigorous eddy-permitting estimates of the global ocean and sea-ice circulations within reach: 1) the conguration of an ecient eddy-permitting global-ocean and sea-ice model that achieves a throughput approaching ten years of model integration per day of computation, 2) the demonstration that boundary conditions estimated at coarse resolution have some skill when applied to an eddy-permitting model, 3) the development of an inexpensive yet eective methodology for calibrating model parameters and for blending estimates from dierent so-lutions and data products, and 4) a hierarchical Kalman lter that can estimate model uncertainties
|
|---|