Water Engineering and Development Centre

Flow structure information from single-point velocimetry


WEDC staff involved: Chris Keylock


Summary of project activities

We have developed a methodology for determining flow structure information from single-point velocimetry; something that his not straightforward without direct access to spatial velocity derivatives.

The technique is based on the joint analysis of a single velocity component and its accompanying Hölder components characterising the intermittency. The theory behind this approach goes back to some qualifiers introduced in to his extended turbulence scaling laws by Andrey Kolmogorov in 1962.

Our approach combines information on the velocity and intermittency using the quadrant analysis framework commonly adopted in boundary-layer analysis. However, out two variables are the fluctuating longitudinal velocity and its intermittency, not the fluctuating longitudinal velocity and the fluctuating vertical velocity. As shown in the figure, this approach can successfully discriminate between various canonical flows, and is also of use for environmental fluid mechanics.


arvindquadrant

Above: The velocity-intermittency structure for a variety of turbulent flows. The four quadrants represent: Q1 – fast, low variability flow; Q2 – slow, low variability flow; Q3 – slow, highly variable flow; Q4 – fast, highly variable flow. The x-axis is a threshold hole size, H, that isolates increasingly extreme events. The y-axis is the probability that the flow is in a given quadrant, with values normalized such that they sum to 1.0 over all four quadrants for all H. The red line is for a turbulent jet, the grey solid and dotted lines are two wake flows at different Reynolds numbers, the solid blue and green lines are a boundary-layer near the wall at two Reynolds numbers, the dotted blue and green lines are a boundary-layer far from the wall at two Reynolds numbers, and the black line is for flow over a set of bed-forms.


Summary of project outcomes and impact

  • Having developed this methodology and applied it to a number of canonical flows as well as turbulent flow over bed-forms, we have also been looking into this as a means to help understand flow through and over plant canopies.
  • In 2018 and 2019, our approach has been used by a group in the USA to characterize the turbulent wakes generated by wind turbines.
  • Work on this topic has led to collaborations with (in chronological order): Nagoya University, University of Oldenburg, University of Minnesota, Simon Fraser University, Duke University, M.I.T., University of Western Australia.

Find out more >

Our original paper on this topic published in Journal of Geophysical Research (2012)

An application to the flow over bed-forms published in Earth Surface Processes and Landforms (2014)

An application using high resolution large-eddy simulation published in Journal of Fluid Mechanics (2016)