Large structure in a turbulent boundary layer

Abstract

The authors claim that their data do not confirm the results of Gordon and Witting [1977], who noted that velocity fluctuations measured in a tidal boundary layer reflected contributions from a large-scale, quasi-ordered motion and a 'background of relatively isotropic turbulence of smaller scale.' Gordon and Witting's observation is similar to numerous observations made in the laboratory turbulent boundary layer. The back or upstream side of the large-scale motion is easily distinguishable by a relatively sudden change [e.g., Chen and Blackwelder, 1978; Phong-anant et al., 1980] in temperature in boundary, layers with neutral and nonzero stratification. This been apphed. In this latter reference, significant correlation was found between the instantaneous shear stress at the smooth wall and u'w' at higher levels; it seems reasonable to expect a similar correlation between the instantaneous boundary shear stress that controls the sediment movement and u'w' at other levels in the river. At least six negative excursions in u'w' of the order of 20-30 u'w', appear simultaneously on all four traces of Figure 15 but the time scale in this figure (and in Figure 14) is inadequate to establish vertical coherence for these excursions and establish a relation with the large-scale motion. A time scale similar to that used in FigUre 16 would sudden change accompanies a rapid change in fluctuations of have been more appropriate for Figures 14 and 15. On the u', w', or the product u'w'. It also seems likely that the bursting phenomenon observed near a smooth wall and other 'events' mentioned by the authors are perhaps best interpreted as the signature of the large-scale motion. It is difficult to determine the precise contribution of this motion to u'w' as this determination would require accurate estimates of both the frequency and of the ensemble averaged distribution of u'w' associated with the large-scale motion. The correlations in Figure 16 are not inconsistent with the notion of a large-scale motion inclined in the direction of tl'ie shear to the river bottom, in as much as u' at higher levels leads u' at levels nearer the river bottom. There is also little doubt that enhanced correlations between u' (or w', u'w') at different levels would be obtained if an enhancement technique similar to that used by Brown and Thomas [1977] had Copyright ¸ 1981 by the American Geophysical Union. basis of the evidence presented, it is difficult to accept the claim by the authors that their data do not confirm the results of Gordon and Witting [1977] and others. REFERENCES Chen, C.-H. P., and R. F. Blackwelder, Large-scale motion in a turbulent boundary layer: A study using temperature contamination, J.

On the Formation Mechanisms of Artificially Generated High Reynolds Number Turbulent Boundary Layers

“…11b for the correlations presented above show 14 • ≤ α ≤ 18 • without any clear trend depending on the trip used. This result is in a good agreement with previous studies such α = 18 • by Brown and Thomas (1977), α = 16 • by Robinson (1986), α = 15 • by Boppe et al (1999) or α = 14 • by Marusic and Heuer (2007). Note that, for brevity, results are only shown for the highest Re, i.e.…”

“…The interaction between the outer and inner regions of the boundary-layers has always been a topic of research interest, for instance the concept of inactive motions of Townsend (1976), some of the results shown by Brown and Thomas (1977) also suggest this kind of interaction, or the mixed inner-outer scaling proposed by DeGraaf and Eaton (2000) among many others. This interest has been intensified in the last few years since the proposal of a modulation mechanism of the inner scales by the outer motions as being responsible for this interaction (Hutchins and Marusic 2007b).…”

Experimental investigation of Reynolds stress complex eddy viscosity model for coherent structure dynamics

“…(10) in the two-dimensional form is expressed as: [2] adopted the constant CEVM, but the eigenfunction showed an oscillatory behavior, which was inconsistent with Brown's experimental results [3]. Falco et al [4] presented a structural model of 'typical eddy' formed on the upstream of large scale motions, and found there was a phase relationship between 'typical eddies' producing most of the Reynolds stress and large scale motions associated with significant Reynolds stress contributions.…”

Phase relations in a forced turbulent boundary layer: implications for modelling of high Reynolds number wall turbulence

“…The endogenous nonlinear forcing arising at the sum wavenumber k 3 from the interaction ofũ 1 andũ 2 is a component of the total forcing f 3 , which results in the velocity responseũ 3 through the transfer function H 3 . The net triadic contributions from the rest of the turbulent scales are lumped together and represented byg 3 . Note that similar interactions betweenũ 2 andũ 3 contribute to forcingf 1 that drivesũ 1 , and interactions betweenũ 1 andũ 3 contribute to forcingf 2 that drivesũ 2 .…”

“…The turbulence signal was split into large-and small-scale components by using a suitable filter, and the large-and small-scale relationship was inferred through studies of superposition effects [1] and correlations [2][3][4]. One of the key observations of these early works was the distinct influence of large scales on the dynamics of large-and small-scale coupling [2].…”