| This dissertation has three parts, two of which are detailed accounts, while the third is a short summary. In the first part, we report the analysis of the three dimensional flow in the inner part of a fully-developed rough channel at Retau=3520, using holographic particle tracking, with the goal of elucidating the formation of coherent structures in the roughness sublayer. We performed experiments in an acrylic rectangular channel, with the top and bottom surfaces containing closely packed, uniformly distributed pyramidal roughness elements, with wall-unit normalized roughness height K+=65 and wavelength lambda=3.2mm. Recording of holograms through the acrylic rough walls was enabled by matching the optical refractive index of the fluid with that of acrylic. We used a 62% by weight solution of NaI in water as the working fluid. In our experiments, we incorporated localized particle injection to ensure that the flow was sufficiently seeded, and this enabled the detection of 5000-10,000 particle pairs in each hologram pair within the 3.1x2.1x1.8 mm3 sample volume, which covered the entire roughness sublayer. Using these particle pairs, we resolved, for the first time ever, the 3D volumetric velocity field within the roughness sublayer of a well-characterized channel flow. First, we assessed the data quality by evaluating how well the data satisfied the continuity equation for varying resolutions and procedures. Mean velocity and Reynolds stress profiles were compared to 2D PIV data, acquired under the same flow conditions by Hong et al. (2011, 2012), showing excellent agreement above one roughness height away from the wall, and discrepancies closer to the rough wall, some of which can be attributed to spatial resolution and bias. Instantaneous realizations indicated that the roughness sublayer is flooded by low lying spanwise and groove parallel vortical structures, as well as quasi-streamwise vortices, some quite powerful, that rise at sharp angles. Conditional sampling and Linear stochastic estimation (LSE) revealed that the prevalent flow phenomenon in the roughness sublayer consists of interacting U-shaped vortices. Such vortices were conjectured in Hong et al. (2012) by using their 2D PIV data. Their low-lying base with primarily spanwise vorticity is located above the pyramid ridgeline, and their inclined quasi-streamwise legs extend between ridgelines. These structures form as spanwise vorticity rolls up in a low speed region above the pyramid's forward face, and is stretched axially by the higher speed flow between ridgelines. Ejection induced by interactions among legs of vortices generated by neighboring pyramids appears to be the mechanism that lifts the quasi-streamwise vortex legs and aligns them preferentially at angles of 54°-63° to the streamwise direction.;In the second part, we describe field experiments involving holography that were performed with a motivation to understand bio-physical interactions in the ocean water column that affect the coastal ecology. A submersible holographic imaging system, "Holosub", was deployed as part of a collaborative effort to characterize the coastal water column in Eastsound, Washington. The Holosub recorded holograms at two magnifications simultaneously, while traversing the water column. Analysis of >20,000 holograms recorded during two independent ascents, provided the undisturbed size and spatial distributions of particles and organisms (identified at least to major taxa), orientation of diatom chains, and the mean shear strain and turbulence dissipation rate profiles. This information was complemented by concurrently measured depth profiles of parameters like density, chlorophyll-a concentration, particle absorption signature and Volume Scattering Function (VSF). Discrete water samples were also collected and analyzed by an on-board flow cytometer, microscope and a high resolution holographic cinematographic setup, while parts of the samples were preserved and analyzed in the laboratory to identify phytoplankton species. Using these data, several key findings were made: (a) a prominent thin layer of non-motile small particles, many of them chlorophyll rich, formed in a region of near zero shear and low dissipation rates, spanning depths just above and within a strong pycnocline; (b) multiple particle concentration peaks formed in other low shear/dissipation regions at several depths; (c) Chaetoceros socialis colonies had several volume fraction peaks in and around the pycnocline, with the largest peak coinciding with the thin layer and some smaller ones coinciding with small particle peaks; (d) zooplankton avoided regions with elevated C. socialis and/or particle concentrations; (e) diatom chains had nearly horizontal alignment (zero angle) in the thin layer, with low angles also observed in other low shear/dissipation regions, while being randomly oriented in several local shear/dissipation maxima regions; (f) size distribution of small particles resembled a power law, while larger diatom chains and C. socialis colonies had a log-normal size distribution.;In the third part, we summarize the investigation aimed to quantify the response of the calanoid copepod, Acartia Tonsa, to dinoflagellates with varying levels of toxicity. The detailed account of this work has been presented in Part II of the dissertation of Hong (2011). The swimming and feeding behaviors of free swimming A. Tonsa were analyzed using high speed cinematic holography, while they were exposed to various mixtures of control food (Storeatula major), and toxic and non-toxic strains of Karlodinium veneficum and Karenia brevis. Quantitative analysis of the copepods' feeding appendage beating revealed two distinct feeding modes: "sampling beating" (short beating duration) and "grazing beating" (long duration). Exposure to mono algal diets containing either of the toxic food strains resulted in low grazing rates. as compared to the control diet (S. major). Exposure to mixtures of S. major and toxic K. Brevis caused a significant reduction in both sampling and grazing, while for mixtures of S. major and K. veneficum, both modes persisted. This difference in behavioral response of A. Tonsa to the different toxic strains was explained based on the documented differences in the effects of brevetoxins and karlotoxins. Furthermore, both grazing modes were found to have lognormal distributions, reasons for which were also explored. |
