Decomposition of velocity fluctuations in the bottom boundary layer

The need of a new decomposition technique for field-measured three-dimensional velocity fluctuations in the bottom boundary layer was identified from an analysis of wave group effects on sediment resuspension and transport. Based on the conceptual framework of a traditional decomposition method, the Linear Filtration Technique (LFT), a new decomposition technique called the Wavelet Linear Filtration Technique (WLFT) was developed for general analysis of the nonstationary behavior of velocity signals due to wave groups.; During the sensitivity tests of these methods, the differential techniques referred to the DLFT and the DWLFT were developed to correct the decomposition problems of the LFT and the DLFT. These differential techniques eliminated several problems of the LFT and the WLFT including directional spreading effects of waves and the three-dimensional nature of the flow field in bottom boundary layer flows. Differential techniques were shown to give reasonable decomposition results to both uniform and turbulence noise signals in the sensitivity tests.; Eight blocks of three-dimensional velocity signals, representing all types of boundary layers, were selected for the decomposition analysis from a total of 660 MB worth of field experiment data. Decomposition results showed that the DLFT and DWLFT produced accurate results in the decomposition of velocity fluctuations, especially in the transverse horizontal, u-direction signal. In the wave dominant horizontal direction and the vertical direction, the LFT demonstrated comparable decomposition results when compared with the differential techniques and exhibited similar results to the spectral splitting technique. All the results clearly demonstrated the advantages of the differential methods in decomposition of field-measured, three-dimensional velocity fluctuations into current, waves and turbulence.; Turbulence statistics computed by the new decomposition techniques made possible the investigation of real three-dimensional turbulence structures. It was shown that the Turbulent Kinetic Energy (TKE,q) is heavily dependent on the strength of the wave, whereas the turbulent Reynolds stress ({dollar}Rsb{lcub}s{rcub}{dollar}) is strongly dependent on the current magnitude and wave-current interaction in wave-current boundary layers. The correlation coefficient ({dollar}Rsb{lcub}s{rcub}/q){dollar} was shown to be an effectiveness indicator of turbulence bottom boundary layers in equilibrium.; The DWLFT method was proven to be superior to the DLFT in the decomposition of nonstationary velocity signals. The decomposition result of velocity signals of block 2196, which was determined to be as a nonstationary forcing period, demonstrated the effectiveness and usefulness of the DWLFT signal in detecting nonstationary wave group influence in the bottom boundary layer velocity signals. The instantaneous temporal variation of q and {dollar}Rsb{lcub}s{rcub}{dollar} from DWLFT signal shows the probable cause of high turbulent diffusion of suspended sediment in the first half of block 2196.