OSNAPi is a conceptual mooring line connecting the German / Canadian western boundary mooring array with the UK eastern boundary moorings. Rather than traverse via east / west Greenland (as for OSNAP), we aim to calculate the central basin transports using a climatology referenced to surface altimetry. It quickly became apparent that existing climatologies did not adequately resolve major currents so I have been developing an improved one, gridding by distance along the OSNAPi line.

Observations consist mainly of CTD data at the continental margins, and Argo data in the ocean interior. I populate an initial month-mean climatology using an expanding search window, and grid in density space. This ‘average month’ climatology forms the basis for the time-varying climatology, by comparing its T and S properties and vertical gradients to the individual data profiles, and warping in 3D (x-distance, depth, time) to try to minimise error. Finally, the gridded profiles are stabilised using the Gibbs Seawater functions.

Argo profiles (blue), World Ocean Database profiles (green) and moorings (red circles) admitted into the QC stage of the climatology
An example of a density-gridded profile from the climatology (black) and the raw Argo and CTD profiles it was derived from (green). Horizontal blue dashed line shows mixed layer depth, averaged from each profile (red points). Horizontal solid red line shows seabed at the profile location (data below this is cut off). Most Argo floats only profile down to 2000 dbar, so these profiles are extended using vertical gradients from more sparse CTD occupations (yellow).
Example time-mean density (Sig-0) section for July
Inferred northward geostrophic transports for time-mean climatology
Initial plots from time-varying climatology, showing the evolution of temperature, salinity and density over time (y-axis).

We are still in the process of validating the climatology, but initial tests are promising. In particular, it is possible to compare the dynamic height inferred from the gridded water properties to the sea surface height obtained from satellite. Even in sparsely-sampled regions this comparison shows good coherence between the two.

Dynamic height inferred from the gridded climatology (red) compared to the sea surface height observed by satellite at this location.