A multi-dimensional spectral description of ocean variability

Ocean current variability at scales other than those of immediate interest limit our ability to identify climate-related trends, make predictions, or test dynamical theories. Multi-dimensional spectra are useful for understanding low-frequency variability, and the spectral shape gives insight into the relevant dynamics. We compute 3-dimensional (zonal/meridional wavenumber and frequency) spectra for the existing TOPEX/POSEIDON-Jason satellite altimetry record. We focus on characterizing the spectral shape and energy containing scales. Westward propagating features with time scales of a few months and spatial scales of order 100 km are dominant throughout much of the global ocean, accounting for roughly half of the variance. While the observed phase speeds roughly agree with those predicted for liner long Rossby waves, the observed dispersion relation, in the form of the zonal wavenumber-frequency spectrum, is inconsistent with the predictions. To explore the vertical structure of the propagating features we study the ECCO2 model, which has an SSH spectrum largely consistent with the altimetric observations. When the vertical velocity structure is decomposed into barotropic and first baroclinic modes, these modes are correlated such that surface currents are enhanced and deep currents diminished. This correlation is increased over rough topography, such as the Mid-Atlantic Ridge, suggesting a role for topography in controlling Rossby wave propagation speeds and vertical structure. In this paper, we present a description of the observed SSH spectrum, the associated vertical structure in the ECCO2 model, and describe the role of large- and small-scale topographic variations in setting the dispersion relation and vertical structure of Rossby waves.

We find that large-scale topographic slopes do not explain the observations, consistent with earlier work. However, small-scale topographic variations do lead to consistently increased phase speeds for Rossby waves as seen in the altimetry, and to enhanced surface/diminished abyssal currents as seen in the ECCO2 model.