A new method to detect mesoscale eddies in satellite records
Event: 2017 Ocean Surface Topography Science Team Meeting
Session: Science III: Mesoscale and sub-mesoscale oceanography
Presentation type: Type Poster
Contribution: PDF file
A significant part of the ocean variability in mesoscale is associated with rings, planetary waves, meanders, and others are known by the generic name eddies. Although this term represents a wide number of different features, here we are focused on mesoscale eddies as fluxes in rotation with spatial scales of hundreds of kilometers and temporal scales from weeks to months. Ocean eddies are particularly important for the transport of momentum, heat, salt and nutrients away from their source because they are patches with properties distinct from that of their surroundings. To clarify which features will be considered eddies in our future analyzes, we defined them as closed instantaneous streamlines roughly circular and relatively persistent. Our proposal is to implement a new methodology to separate the signal of nearly-circular mesoscale eddies that are ubiquitous in a variety of radiometer and altimeter records. The proposed method uses an algorithm based on the Radon transform and its inverse function to detect areas occupied by these eddies. The basic premise of this study is that the inverse Radon transform of a circularly symmetric feature is angle-independent. Therefore, the more constant the Radon transform of an image is, the more circularly symmetric the image is. We are able to quantify how circular a feature observed in a satellite image is and suggest whether it is an eddy or not, based on a geometric criteria. A series of tests with synthetic data were performed to assess the sensibility to spectral content, and aspect ratio detectable by the method. Another sequence of tests was carried with realistic sea surface height data. In this last one five areas of 100 by 50 pixels were selected and filtered with exactly the same algorithm and parameters. The tests were expanded to include data sets of sea surface and chlorophyll-a concentration. All the set of tests indicate that the proposed method was able to identify and track eddies in a single time series of satellite data and in multiple series of different variables. The main advantage of this method in comparison with other automatic ones based on dynamics and geometry or manual methods is its versatility to detect eddies in multiple data sets of different variables. Another point is that, eddies tend to preserve their circular symmetry for dynamical reasons and this is the heart of the proposed method.