Bursting phenomenon in turbulent wall-bounded flows

The scaling of bursting period governing the near-wall turbulence production in a wall-bounded flow has been an unresolved issue in the literature, despite nearly half a century of intense research. By using measurements from a turbulent channel flow along with laboratory boundary layer and atmospheric boundary layer data available in the literature, the appropriate scaling for the inter-burst time period (time period between two consecutive bursting events) is reexamined. The bursting period, non-dimensionalized using the inner scales, varied by five orders of magnitude over the Reynolds number range of 10(3)-10(7), conclusively demonstrating thereby that the bursting period does not scale on inner variables. The bursting period, non-dimensionalized using the traditional outer timescale (ratio of the channel half-height to the centerline velocity), was found to asymptote to an invariant value at high Reynolds number for laboratory boundary layer and channel data. However, when the atmospheric boundary layer data were also included, this non-dimensional timescale's invariance became weaker and showed a slowly increasing trend with Reynolds number. A new modified outer timescale, where the friction velocity is the velocity scale, resulted in invariance for Reynolds number greater than 900 and suggests possible invariance over a wider range extending to higher Reynolds numbers. The new timescale is also amenable to a simple physical interpretation as the time of transit of the up-welling eddy to reach the outer region of the boundary layer/centerline of the channel flow.