On intraseasonal variability in the tropics: tropical cyclones, the Madden-Julian Oscillation, and equatorial waves

Abstract

This dissertation addresses some aspects of tropical intraseasonal variability, which is dominated by tropical cyclones, equatorial waves, and the Madden-Julian Oscillation (MJO). Because these phenomena have significant societal impacts on sub-seasonal time scales, and it is important to understand how they interact with the large-scale atmosphere. The first part of this thesis develops Forecasts of Hurricanes using Large Ensemble Output (FHLO), a large-ensemble, probabilistic tropical cyclone forecast model. FHLO incorporates the state-dependent forecast uncertainty by sampling the internal variability of ensemble numerical weather prediction models. It is shown that including state-dependent forecast uncertainty can lead to significant improvements in pointwise wind speed forecasts on lead times longer than around 3-days. The second part of this thesis addresses how tropical disturbances interact with the stratosphere. A linear framework in which a convecting, quasi-equilibrium troposphere is coupled to a dry, passive stratosphere is developed. It is shown that smaller scale waves are strongly damped by the stratosphere, while slower propagating waves, such as Rossby waves and the MJO, are less affected by the stratosphere. Excitation of the barotropic mode by the stratosphere and surface friction is also analyzed. In particular, it is found that surface friction can excite the barotropic mode far away from the equator, though the poleward extent of the barotropic mode is strongly controlled by how much energy leaks into the stratosphere. The last part of this thesis extends the linear framework to include non-zero zonal wind in the stratosphere, to understand how stratospheric circulations, like the Quasi-Biennial Oscillation, can influence the strength of the MJO. It is found that the tropospheric barotropic mode can be phase-shifted by stratospheric winds, but only under unrealistic forcings at the tropopause. Upward wave radiation is found to be stronger under easterly than westerly winds in the stratosphere, because of increased upward energy flux by Kelvin waves. The effect of the stratosphere on cirrus clouds is also investigated. It is shown that dynamical modulation of lower stratospheric clouds, and anomalous advection of upper-tropospheric ice clouds, can explain why the MJO is stronger under easterly than westerly phases of the QBO.