Predictability of the quasi-biennial oscillation driven by land-sea breezes.

• Main Author: Wan, Fang
• Other Authors: Koh Tieh Yong
• Format: Theses and Dissertations
• Language: English
• Published: 2014
• Subjects: DRNTU::Science
• Online Access: http://hdl.handle.net/10356/55454
• Institution: Nanyang Technological University

This dissertation studies the predictability of the quasi-biennial oscillation (QBO) driven by land-sea breezes. Particularly, the proposed subject is approached through an idealized numerical model of the QBO driven by land-sea breezes, and analysis of separation-growth dynamics of the idealized numerical model. Implicit to the definition of separation growth is the fundamental concept of a metric, i.e. the distance between two states in the phase space. We suggest that some energy metrics are based on linearization and hence it may not be appropriate to apply these metrics when two states are far apart, such as encountered in the real atmosphere. A new approach to define the energy metric is proposed and it is used to study the separation-growth dynamics in the idealized model. We firstly derive the new separation metric based on energy for atmospheric models of various levels of sophistication: 2-D barotropic model, shallow water model, 3-D compressible dry model in different vertical coordinates. Atmospheric phase vector is defined such that square root of the energy is the Euclidean norm. Numerical calculations of the new energy metric are illustrated with analytical dynamic systems as well as with global reanalysis data. Linearized separation metric in more conventional basis is found to have nonquadratic terms, which are not negligible in nonlinear fluid. It is also found that compare to our proposed separation metric, the total energy norm in literature has doubled the contribution from enthalpy separation by having a multiplication factor of two. Besides, it underestimates the surface pressure separation by one order of magnitude. Secondly, with the aid of state-of-the-art numerical weather prediction model, we successfully simulate a QBO-like oscillation of the zonal mean flow driven solely by land-sea breezes. The 2-dimensional QBO model resolves the waves explicitly. And compared to the general circulation models (GCMs), the computational cost of our model is moderate, which allows us to run ensemble experiments with long integration period. The simulated QBO propagates from lower stratosphere to lower troposphere with symmetric westerly and easterly regimes. The simulated QBO period is 440 days. Lastly, we apply the new metric to study the separation-growth dynamics of the simulated QBO driven by land-sea breezes. The separation growth is found to be insensitive to some characteristics of the random initial perturbation. The simulated QBO and land-sea breeze in our model can be regarded as climate and weather variations respectively according to their time scale. The weather variations are found to be chaotic as the initial separation grows exponentially mainly due to the afternoon convection. And in a few days, the separation reaches a saturation level, which is independent of the initial perturbations. However, the climate variations which are defined as the smoothed zonal mean variables are found to be rather regular and predictable. The zonal mean wind in the perturbed run is found to be almost the same as that in the control run within 1000 days.