To permit the tracking of turbulent flow structures in an Eulerian frame
from single-point measurements, we make use of a generalization of
conventional two-dimensional quadrant analysis to three-dimensional
octants. We characterize flow structures using the sequences of these
octants and show how significance may be attached to particular
sequences using statistical mull models. We analyze an example
experiment and show how a particular dominant flow structure can be
identified from the conditional probability of octant sequences. The
frequency of this structure corresponds to the dominant peak in the
velocity spectra and exerts a high proportion of the total shear stress.
We link this structure explicitly to the propensity for sediment
entrainment and show that greater insight into sediment entrainment can
be obtained by disaggregating those octants that occur within the
identified macroturbulence structure from those that do not. Hence, this
work goes beyond critiques of Reynolds stress approaches to bed load
entrainment that highlight the importance of outward interactions, to
identifying and prioritizing the quadrants/octants that define
particular flow structures.
Key Points
<list list-type=''bulleted'' id=''jgrf20196-list-0001''>
<list-item id=''jgrf20196-li-0001''>A new method for analysing
single point velocity data is presented <list-item
id=''jgrf20196-li-0002''>Flow structures are identified by a
sequence of flow states (termed octants) <list-item
id=''jgrf20196-li-0003''>The identified structure exerts high
stresses and causes bed-load entrainment