| dc.description.abstract | The Asian summer Monsoon circulation is driven by differential thermal heating, primarily associated with the localized latent heat release from enhanced precipitation over the India sub-continent. Although this heating is of limited zonal extent, it drives a time-averaged, upper level anticylone which is of global extent, extending from the western edge of the bulge of Africa, to the east of the Asian continent. The current theory (originally proposed by Gill (1980)) for explaining this zonally asymmetric component of the tropical circulation is unsatisfactory because it is based on the linear theory of damped equatorial waves while it is known that, at least for the upper level flow near the tropopause, the dynamics are strongly nonlinear. An alternative explanation, which is consistent with the nonlinear nature of the flows, involves the shedding of vortices from the directly forced monsoon anticyclone. The vortices, or eddies, are capable of drifting to the far field to establish a circulation which extends far beyond the local forcing. This thesis provides a dynamical explanation for the generation of eddies near the center of a divergent anticyclone, which, through their westward drift are responsible for the establishment of the global scale of the Asian summer Monsoon. The thesis consists of two parts, one numerical and one observational. The numerical study systematically investigates localized thermally driven circulations by using a shallow water model. This part of the thesis is theoretical in nature, and seeks to understand how non-axisymmetric elements such as a beta effect, or an external uniform flow, affects the dynamics of a divergent anticyclone for which, in the absence of non-axisymmetric elements, there exists an analytical axisymmetric solution. Control parameters which determine the dynamical regime of the flow are identified and explained. For the midlatitude beta plane experiment, the control parameter, pO, is the ratio between the free drift speed of an axisymmetric vortex on a beta plane, OL, and the strength of the forced localized divergent flow (ux) where L is the size of the axisymmetric circulation. For the uniform flow experiments, the control parameter is the strength of the uniform flow, Urn, and the divergent flow, ux. Each control parameter measures the relative importance of two competing effects, one which tries to displace the anticyclone westward (for the midlatitude beta plane experiments), or downstream (for the uniform flow experiments) and one tries to keep the forced vortex anchored. For each series of experiments, a critical value which separates the different long-time flow behavior is found. When the circulation is below the critical value, the circulation is persistent and localized. When the control parameter is above the critical parameter, a material filament with low potential vorticity is drawn from the divergent center, rolls up due to shear instability and is soon shed away by detaching itself from the main vortex. In the time-mean vorticity budget, the transient eddies have the effect of dissipating the time-mean flow. The dissipation effect by transient eddies can be grossly parameterized as a linear damping term in the linear version of the model. Another series of experiments, extending the midlatitude beta plane to the equatorial beta plane, with an equator within the reach of the forced perturbation, is conducted in which equatorial waves can be generated. The shedding behavior begins when the value of the control parameter is of order unity in the midlatitude beta plane experiments, and continues to exist for values of order 10 in the equatorial beta plane experiments. When the control parameter takes on values of order 100 and larger, the shedding behavior disappears and is replaced by linear wave solutions. In these experiments, another non-dimensional parameter (q), which measures the non-dimensional distance from the thermal forcing center to the equator, is found to affect the stability characteristics of the forced vortex. This series of experiments also allows equatorial waves to co-exist with the nonlinear vortex and to be excited by the broad thermal cooling, whose magnitude and location are determined by the internal dynamics of the nonlinear forced vortex. This linear part of the response is similar to the solution predicted by Gill (1980), but is of opposite sign, since it is the response to the resulting broad thermal cooling part of the thermal forcing, and not the small localized imposed thermal heating as Gill would have it. In the second part of the thesis, the theory is confirmed by discovering eddy shedding from the analysis of observational data. The potential vorticity on the isentropic surfaces are analyzed from 17 isobaric level NCEP-reanalysis data over the region of the Asian summer Monsoon. Two episodes of eddy shedding are found in July of 1990. The shedding events in the potential vorticity field are observed at the levels of isentropic surfaces from 360K to 380K. The induced geopotential perturbation penetrates deeper to 400 mb. The technique of Contour Advection with Surgery, a technique that allows to discriminate between adiabatic and diabatic effects, is used to recapture the shedding events, and confirm that the eddy shedding is indeed due to the essentially inviscid process identified in the idealized shallow water model. | en_US |
