Introduction: Convection is one of the major modes of heat transfer in fluids and is especially important given the large scales of most geophysical flows. Heat transfer by natural convection plays a significant role in the structure of Earth's atmosphere, oceans and mantle. Flow in a horizontal layer of fluid heated uniformly from below and cooled from above—Rayleigh-Be´nard convection—is an idealized problem from which much can be learned about the nature of convective flow and heat transfer in physics, astrophysics and geophysics. On the other hand, the flows resulting from a horizontal difference in temperature or heat flux at a single horizontal boundary of a fluid — ‘horizontal convection’— is a basic model for thermally driven ocean circulation. Therefore, knowledge of these processes is essential to understand their consequences for global climate.
GFD Contributions: We are investigating the role of convection in contributing to forcing or feedback in the ocean circulation. Group research focuses on:
Laboratory experiments: We use laboratory apparatus to examine the underlying physics of horizontal convection, and to obtain insights into its relevance to ocean circulation. We consider different geometries, such as a rectangular box (ocean basin) and an annulus (Southern Ocean), placed on a rotating table to model the effects of Earth's rotation. A gradient in heat or salinity is applied at either the upper or lower horizontal boundary to simulate heating and cooling at the ocean surface – this gradient in surface density causes convective circulation. We are also studying the convective flow at an ice-seawater interface and the rate of ice melting under Antarctic ocean conditions. We use a variety of techniques to quantify the flow, including thermistor temperature probes, conductivity probes, digital photography and particle tracking velocimetry.
Numerical simulation and modelling: Numerical simulation is a great tool to study thermal convection. Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES) are used to obtain solutions under both laboratory and ocean conditions that can be compared with experimental data, and they provide a complete view of the mechanical energy budget for both Rayleigh-Be´nard and horizontal convection under strong forcing (high Rayleigh numbers). Large Eddy Simulation (LES) will potentially enable the study of convection to extend into the asymptotic (‘ultimate') regime at very large Rayleigh numbers. The simulation of convection can also be combined with other oceanic phenomena, including at a relatively small scale, the melting if Antarctic ice shelves and glacier tongues that involves convection at the ice front, or at much larger scales with mesoscale eddies, Rossby waves and wind-driven circulations.