Estimating diffusion coefficients from ocean hydrography

Supervision Team

Background

A prominent use of ocean models is to predict future climate change under various greenhouse gas scenarios.  This is done using coupled ocean-atmosphere-ice models.  A key parameter in such models is the strength of ocean mixing. Ocean mixing controls the rate at which heat and CO2 are absorbed by the ocean. These models need the strength of ocean mixing processes to be prescribed.

As ocean models have improved over past decades, their ability to accurately model the present ocean is being more obviously limited by the realism of the imposed mixing coefficients.  For example, vertical mixing is commonly set as a constant, or near constant parameter despite it being known to be highly spatially variable. What are lacking are global estimates of rates of vertical and lateral mixing.

To overcome the lack of direct mixing observations we have recently developed a new inverse method; the Tracer-Contour Inverse Method, which is able to deduce the strength of both the vertical and lateral mixing in the ocean, as well as giving superior estimates of the mean ocean circulation (Zika et al. (2009)).  This Tracer-Contour Inverse Model is much more skilful than the three prior inverse methods that have been used in oceanography since 1978. 

This inverse method uses the hydrographic data that is becoming available from the Argo floats; autonomous floats that are providing oceanographers with more data than has been available to date from research ships, although the ship-derived data is vital below 2000m depth. 

This PhD topic will develop the Tracer-Contour Inverse Method (TCIM) so that it can be applied in a general way to any region of the world ocean, and conduct inversions of an isopycnally averaged hydrographic climatology systematically around the world ocean.  The global application of the TCIM will provide an understanding of the spatial variation of the vertical and lateral ocean mixing that are appropriate to be used in coarse-resolution ocean models.  Simple spatial functions of the magnitude of these diffusivities, accompanied by the uncertainties of the estimates, can then be included as mixing parameterizations in climate models. This will improve the model simulation and prediction of the ocean circulation and change.

This research uses the new ocean observing infrastructure (Argo floats) and a new inverse method technology to obtain estimates of ocean mixing, estimates that are needed by ocean modellers for the purpose of climate prediction.

References:
Zika, J. D., T. J. McDougall and B. M. Sloyan, 2009: A Tracer Contour Inverse Method for estimating ocean circulation and mixing.  in press  J. Physical Oceanography.