| I present laboratory experiments to elucidate the fundamental physics of interfacial transport under the influence of turbulence. Such turbulence can come from many sources; I consider the case in which it is generated far beneath the interface and displays high Reynolds number, horizontal homogeneity, low mean flow, and near zero mean shear. This is accomplished by designing, characterizing, and optimizing a novel type of turbulence tank, driven by the Randomly Actuated Synthetic Jet Array (RASJA). This is a planar array of synthetic jets, firing upwards in a spat iotemporally random pattern. In designing the RASJA, I explore the effects of different spatiotemporally random driving patterns, highlighting design conditions relevant to all randomly forced facilities.; In this new apparatus, I use quantitative imaging to directly measure the spatiotemporal dynamics of turbulent flux near the interface. The novel results from this study rely on a quantitative imaging of the dissolved carbon dioxide gas concentration field. This is accomplished by developing a method for calibrating a pH-sensitive fluorescent dye for use as a carbon dioxide tracer. Combining this laser induced fluorescence (LIF) technique with particle image velocirnetry (PIV), I obtain direct measurements of instantaneous flux fields.; Image sequences and statistics from this dataset allow broader understanding of the turbulent flux dynamics. Specifically, a quadrant analysis of fluctuating velocity and concentration fields shows that upward-going and downward-going fluid both contribute equally to the total flux. This demands a reconsideration of the traditional surface renewal theories, and is necessary for extending measurements in simple laboratory tanks to the more complex flows found in natural water bodies. Profiles of variance, wavenumber spectra, frequency spectra, and structure functions near the interface show variation with depth and may aid in the interpretation of field data.; The RASJA allows us to perform several additional measurements. I study a model of the Eulerian frequency spectrum that gives the dissipation rate from velocity timeseries at a single point without the use of Taylor's frozen turbulence hypothesis. I also quantify the effects of the mean flow in stirred tanks, reporting how it affects the rate of gas transfer across the free surface. |
