Previous studies pointed out that due to the acceleration or deceleration action of tide current, the flow structure deviates from traditional logarithmic law in estuary, coast or other near shore water. The tidal velocity distribution model was derived and compared with the traditional logarithmic model. It should be pointed out that the velocity data adopted have four layers within one meter above the bed, and the roughness length z0 is different in the two models even in the same velocity profile. Because the fluctuation of roughness length z0 is remarkable when determining by single velocity profile, some studies thought that the variation of roughness length was small between adjacent time when the change of topography was less obviously. Therefore, the measured data is divided into several sections by one day or a tidal cycle to fit the velocity profile of every section to obtain a roughness length z0, i.e., the roughness length z0 varies only after a day or a tidal cycle. The purpose of the paper is to expand the log-linear model to full depth by adding the surface boundary condition
uzz=D=0(Diswaterdepth)
and to discuss the difference when 6 points (bottom layer, 0.2D, 0.4D, 0.6D, 0.8D, surface layer) velocity profile are fitted by logarithmic model, log-linear model, and extended log-linear model with the same roughness length z0 in different time section, respectively. The calculated friction velocity and friction coefficient and their correlation are discussed. The results show that the log-linear model and the log-linear extend model are closer to the measure velocity profile than that of the logarithmic model.
This content is only available via PDF.
You do not currently have access to this content.