Code for the detection and tracking of eddies, following Chelton et al. (Prog. Ocean., 2011) given a series of sea level maps.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Automated eddy detection

Eric Oliver

Code Description

File Description Code for the detection of eddies given a series of sea level maps Code for the tracking of eddies after detection has been performed Parameter file used by eddy detection and eddy tracking programs Added by Chris, support file for for NEMO output to calculate how many timesteps in an experiment Code for calculating census statistics of tracked eddies Code for plotting eddy tracks Module of supporting functions


  1. This code as been applied to model output from OFAM and from weekly and daily sea level maps from Aviso. To apply it to another dataset it is necessary to make the following adjustments:

a. Add a new dataset id 'NAME' in the conditionals in, including relevant parameters (number of time steps, resolution, time step).

b. Add appropriate code in 'load_lonlat' and 'load_eta' functions (both in to properly handle the loading of your data. Code assumes one file (spatial map) per time step.

  1. rosrad.dat obtained from here. Specifically:

Changes added by Christopher Bull (Dec 2015)

Modified to work with NEMO. Made some general changes including:

  1. Works with python netCDF4 library.
  2. Refactored so global data and plot dirs are defined in (and created if they don't exist)
  3. Refactored so that pathroot for input data can be defined in params rather than eddy_functions.
  4. Added quick_plot and detection_plot functions to
  5. Added functions from ecoliver (that were not previously included) into
  6. Added rossrad.dat


Eric C. J. Oliver
Institute for Marine and Antarctic Studies
University of Tasmania
Hobart TAS, Australia e:

Christopher Bull. Climate Change Research Centre and ARC Centre of Excellence for Climate System Science. University of New South Wales
Sydney, NSW, Australia, 2052
w: t: @ChrisBullOceanO