| 總結: | <p>This thesis describes the development of and results from a new numerical model of Jupiter's atmosphere. The model is an adaptation of a general circulation model (GCM) of the Earth's atmosphere. It is a high resolution, limited-area model of the upper troposphere and lower stratosphere of Jupiter, with realistic stratification and simple parametrizations of appropriate physical processes.</p> <p>We have used the model to study two areas of jovian atmospheric dynamics: firstly, mid-latitude vortices (exemplified by the Great Red Spot) and their interaction with baroclinic and barotropic jetstreams, and secondly, the properties of equatorial waves and their effect in driving a superrotating equatorial jet.</p> <p>Before describing the main GCM, we present a time-independent model of a balanced axisymmetric vortex. The model performs a potential vorticity inversion, predicting the vertical variation of winds and temperatures required for balance, given a specified PV anomaly. We show that both shallow eddies (confined above about 20 bars) and deep ones (extending barotropically below 100 bars) could theoretically exist. The two types would be almost indistinguishable if observed only above the cloud-tops, i.e. at pressures <em>p</em> ≤ 300 mb.</p> <p>We then describe the primitive-equation GCM, concentrating on the modifications necessary to represent Jupiter. In order to run the GCM, basic-state temperature and zonal wind profiles must be chosen, for these are not adequately constrained by observations. Numerical solution of the vertical structure equation allows the calculation of the eigenmodes of any atmosphere, thus predicting the approximate vertical structure of Rossby waves. This simplified 1D analysis gives an insight into suitable initial profiles and the results of the full model.</p> <p>The first GCM experiments examine midlatitude vortices in the absence of a background flow. Anticyclonic vortices migrate equatorward, while shrinking and intensifying by a process akin to frontogenesis. After intensifying, some eddies become unstable and disintegrate, generating daughter eddies. All eddies seem to adjust their vertical structure toward the first baroclinic normal mode. When eddies are placed in an enviroment of latitudinally alternating zonal jetstreams, meridional migration is blocked by the jets. However, most eddies intensify until they overcome the barrier presented by the jet. Mixed baroclinic/barotropic instability in the jets can generate eddies of either shallow or deep structure, echoing the results of the PV inversion model.</p> <p>Moving on to equatorial dynamics, we analyse the properties of equatorial waves on Jupiter, and their interaction with the zonal mean flow. We find that mean-flow acceleration by dissipation of upward propagating Kelvin waves can (in principle) quantitatively account for the observed superrotating equatorial jet. By considering the effect of a range of possible wave sources below the clouds, we find that the troposphere filters waves primarily by their phase speed, suggesting that the speed and shape of the jet may be less sensitive to the source than to the basic atmospheric stratification. We propose a mechanism of Kelvin-wave forcing, plus weak zonally symmetric solar heating to account for the observed velocity and meridional structure of the equatorial jet. The vertical structure is less realistic, with an underlying easterly jet, but this may be because our sources are insufficiently deep in the atmosphere.</p>
|
|---|