Atmosphere-ocean and atmosphere-land interactions are important processes which determine the development of monsoon systems. In this study, a new atmosphere-ocean surface interaction scheme, referred to as AOSIS, is developed and verified with observed data. AOSIS, together with ALSIS (Atmosphere-Land Surface Interaction Scheme), is then coupled into CEMSYS4 (Computational Environmental Modelling System) to investigate the influences of atmosphere-ocean and atmosphere-land surface interactions on the Australian Monsoon, especially the monsoon onset, break and withdrawal. Numerical experiments are carried out and the simulations are compared with the NCEP (National Center for Environmental Prediction, America) data. AOSIS is constructed with three basic components, i.e., a two-layer ocean temperature model, a wind-wave model and a surface flux model. We divide the ocean into a mixed layer and a deep layer. However, the depth of the mixed layer is not constant but varies with time, depending on surface wind shear and buoyancy flux. In AOSIS, we adapted the approach of relating the stages of wave development by wave age and proposed a new expression for calculating the ocean surface roughness length, $z_{0m}$, with consideration of waves. We test AOSIS in a stand along mode against the Moana data and the NCEP data. The comparison with the Moana data shows that AOSIS has considerable skill in simulating SST (sea surface temperature) and energy fluxes, with the simulated values in good agreement with observed data. AOSIS is also successful in simulating the warm and cool effects considered in the COARE (Coupled Ocean-Atmosphere Response Experiment) scheme. Comparison with the NCEP data also confirms that AOSIS simulates SST well. AOSIS and ALSIS are then coupled into CEMSYS4. We apply the system to the simulation of SST and surface energy fluxes over the Australian region and compared the results with the NCEP data. It is found that the simulated SST and energy fluxes are in good agreement with the NCEP data. Further, we study the synoptic events of the Australian Monsoon onset, break and withdrawal and examine the impacts of atmosphere-ocean and atmosphere-land surface interactions on such synoptic events.

Measurements of the fluctuations of humidity, temperature and velocity were made in the marine boundary layer. The humidity fluctuations were measured with a Lyman-alpha humidiometer. Temperature fluctuations were measured with a dry thermocouple and a platinum resistance thermometer. Velocity fluctuations were measured with a three component sonic anemometer. These measurements were made from the Floating Instrument Platform (FLIP) operated by the Scripps Institution of Oceanography near San Diego in February, 1969 and during the Barbados Oceanographic and Meteorological Expedition (BOMEX) in May 1969. The data were processed by digital techniques and the various spectra, cospectra and quadspectra between the velocities, humidity and temperature were obtained. Integrals of the cospectra were produced which allowed estimates of the fluxes of latent and sensible heat to be made. The normalized spectra of humidity fluctuations in San Diego and BOMEX have similar shapes. The normalized cospectra between vertical velocity and humidity in San Diego and BOMEX have similar shapes. Universal forms for the normalized humidity spectrum and the normalized cospectrum between vertical velocity and humidity may exist. The normalized spectra of the temperature fluctuations in San Diego and BOMEX have different shapes. The differences in shapes may be related to stronger radiation effects during BOMEX than in San Diego. The normalized cospectra between vertical velocity and temperature in San Diego and BOMEX also have different shapes. It is probable that a universal form does not exist for the normalized temperature spectrum or for the normalized cospectrum between vertical velocity and temperature. Directly measured values of the latent and sensible heat fluxes were used to test the validity of the bulk aerodynamic method of predicting the latent and sensible heat fluxes from the mean wind speed and mean air-sea humidity or temperature differences. The limited results from San Diego indicate that the sensible heat flux may probably be predicted from the mean wind speed and the mean air-sea temperature difference in temperate regions. The bulk aerodynamic method was not useful for predicting the sensible heat flux in BOMEX. The observed values for the sensible heat flux were much larger than would be predicted. The latent heat flux could be predicted from the mean wind speed and the mean air-sea absolute humidity difference with a probable error of less than 20%. The validity of the formula developed by Bowen (1926) for predicting the Bowen ratio (sensible heat flux/latent heat flux) was tested with directly measured values of the Bowen ratio. It was found that in San Diego the Bowen ratio could be predicted with a probable error of 15%. The Bowen ratio predicted for BOMEX was too low by a factor of two or more. The ability to predict the Bowen ratio from the ratio of the temperature fluctuations to the humidity fluctuations in the 0.05-0.1 Hz range was investigated. The method predicted the Bowen ratio with a probable error of 15% in San Diego and 20% in BOMEX. / Graduation date: 1971

Graduation date: 1970

The dynamical role of mesoscale eddies in the ocean general circulation is investigated using eddy-resolving quasigeostrophic (QG) and primitive equation (FE) models which are parametrically identical. The results of both QG and FE numerical experiments in mid-latitude, rectangular ocean basins are systematically intercompared and extensively analyzed in terms of basic quantities: energetics, relative and potential vorticity, and eddy momentum and heat transports. Although overall the analyses show that the results are fairly similar between the two models, a closer examination reveals some significant differences. Most of these differences are due to the presence of Kelvin waves along the lateral boundaries of the PE model. These waves are the main source for mean and eddy divergent kinetic energy. Further model parameter studies are needed to determine whether the presence of these Kelvin waves is due to numerics, physics, or a combination of the two, and if the two-day sampling rate commonly used for obtaining eddy statistics significantly aliases these high frequency waves. / Graduation date: 1984

Thesis (Ph. D.)--University of Washington, 2001. / Vita. Includes bibliographical references (leaves 114-120).

Thesis (Ph. D.)--University of Washington, 2003. / Vita. Includes bibliographical references (p. 77-80).

Thesis (Ph. D.)--University of Oklahoma, 1998. / Includes bibliographical references (leaves 215-221).

Thesis (Ph. D.)--University of Washington, 2002. / Vita. Includes bibliographical references (leaves 115-118).

Graduate

No description available.