Physics-Dynamics Coupling: Understanding Tropical Processes and Process Interactions in Weather and Climate Models

Abstract

The tropics are an important area of uncertainty in weather and climate models with variation between models in the tropics often being larger than in the extra-tropics for example due to a lack of strong geostrophic balance. In this thesis, we consider atmospheric behaviour in the tropics with a focus on understanding the long-term balance regimes that arise when drag and heating interact in the tropical atmosphere.

We will look at the dry and moist case as well as considering the adjustment to balance process. We outline a number of long-term balance regimes for the 2D dry and moist tropical atmosphere in the absence of Coriolis acceleration with heating and drag physics, deriving scalings for horizontal velocity, vertical velocity, potential temperature, Exner pressure, and where appropriate buoyancy frequency and total water. We then investigate the ability of a model to achieve or not achieve our hypothesised balances. We find in the dry case that the four regimes we outline are achievable within the model and produce distinct behaviours with scaling relationships that mostly match the theory. In the moist case however, only one balance regime can be achieved by the model, but the existence or not of long-term balance in the moist case has a notable effect on the triggering, sustaining and organisation of moist plumes.

We also look at numerical constraints affecting the development of long-term balance regimes and we find that the horizontal gridlength has a strong effect on which long-term balance regime the model finds itself in and whether the model can achieve balance at all. We also consider the effect of horizontal gridlength on adjustment processes such as gravity waves.