Physical Analysis of Gridded Ocean data

Inter-comparison of model and gridded observations of the ocean is a challenge when they use different grids. Interpolation from one grid to another brings eventual errors that may affect significantly large scale budgets of tracers (heat, salt, freshwater). This suite of programs offers the possibility to analyze gridded ocean data along physical sections with minimum interpolation. For example, this allows to monitor the circulation across an observed array in various model outputs, whatever their spatial resolution or type of discretization (B- or C- grids). When defining sections that enclose a specific volume, large scale budgets of tracers can be reconstruct and inter-compared among all kinds of gridded ocean data.

The core of PAGO programs consists of finding the suite of west and north grid faces to go from one geographical landmark to another, following the geodesic distance. Temperature and salinity are averaged from the centers of the 2 adjacent grid cells to the center of the grid faces (taking into account missing values in land). Velocities, if available, are either located at the center of the grid faces (in case of C-grids), or at the corners (B-grids). In case of C-grid data, no interpolation is required for the velocities. In case of B-grid data, the volume transport at the corner of the grid cells is computed and then split in 2 to the center of each adjacent faces (see more details in the models page).

Here is the list of the main PAGO functions to be called:

  • sections_MODEL reads data regarding grid characteristics only, selects the region of interest when uploading the data, and identifies sections and areas on which circulation, and tracer content will be diagnosed.

  • loaddata_* uploads the data output and extracts the information required along preselected sections and areas. This function also interpolates velocity and tracer data at the center of west and north grid faces, when needed.

  • indices_MODEL calculates simple or full diagnostics of the circulation (transport of volume, heat, salt and freshwater) across selected sections at each time step.

  • volumes_MODEL calculates thermal, haline and freshwater content within selected areas at each time step. It also calculates advective convergence of tracers at the boundaries or areas.


    In order to make it easier for a new user to become familiar with this suite of functions, there are few ready-to-use MATLAB scripts that contain the full list of steps from the definition of the sections to the indices and the mapping of the diagnostics (script_PAGO_*).

    The development of this package is still ongoing. Please inform us of any bugs or necessary improvements. Users of the package may acknowledge this website and the following paper:

    J. Deshayes, R. Curry and R. Msadek (2012): CMIP5 model inter-comparison of freshwater budget and circulation in the North Atlantic, Journal of Climate (submitted).