Dr Chris Wilson (NOC, Liverpool)

Challenging existing models of ocean circulation with Lagrangian insights

The majority of established ocean circulation models (conceptual, theoretical, computational) are built on an Eulerian view : a system evolving in time at a set of fixed points in space.  Computational models of climate are built from a set of governing equations on a fixed, Eulerian spatial grid.   Many textbook examples of conceptual/theoretical models of climate-related processes are built on averages of Eulerian observations or output from Eulerian computational models.   Three such examples include the Transpolar Drift in the Arctic, the Atlantic Meridional Overturning Circulation (AMOC) and the Gulf Stream/North Atlantic Current.  Each of these is responsible for significant transport of heat, salt, momentum and carbon on a range of space and time scales.  All three play a significant role in affecting climate and weather in Europe and globally.   However, Eulerian observations of the ocean, used to constrain models, are very limited and are dominated by observations of the ocean surface only, within the satellite era, augmented by sparse mooring arrays and ship transects to sample the deep ocean.  The last 20 years has marked a transition, with Lagrangian observations being added for the upper 2000 m due to Argo profiling floats, which drift with the flow and sample in the vertical every 10 days.  Perhaps the Argo era and the trend towards increasing use of autonomous (drifting, Lagrangian) observing platforms and reduction in ship-based ocean observation will motivate renewed interest in understanding the ocean from the Lagrangian perspective.

In three parts, this talk will cover the examples above, each with different Lagrangian methods used to test the Eulerian view.   Flow structure and coherence, pathways and bifurcation, predictability, circulation timescales, material transport and mixing properties all form part of the story.  The insights all share a common theme – that the underlying key processes are not as smooth or steady as their canonical Eulerian models would have us believe – but they are also not totally noisy or random – coherent flow structures play a central role, shape the transport and predictability properties relevant to climate, and underly a new and major challenge for climate modellers.

Getting here