Open AccessTurbidity current with a roof: Direct numerical simulation of self‐stratified turbulent channel flow driven by suspended sediment
Author(s) -
Cantero Mariano I.,
Balachandar S.,
Cantelli Alessandro,
Pirmez Carlos,
Parker Gary
Publication year - 2009
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2008jc004978
Subject(s) - turbidity current , turbulence , settling , mechanics , reynolds number , stratification (seeds) , geology , open channel flow , stratified flows , eddy , sediment , reynolds stress , stratified flow , physics , geomorphology , thermodynamics , sedimentary depositional environment , structural basin , seed dormancy , germination , botany , dormancy , biology
In this work we present direct numerical simulations (DNS) of sediment‐laden channel flows. In contrast to previous studies, where the flow has been driven by a constant, uniform pressure gradient, our flows are driven by the excess density imposed by suspended sediment. This configuration provides a simplified model of a turbidity current and is thus called the turbidity current with a roof configuration. Our calculations elucidate with DNS for the first time several fascinating features of sediment‐laden flows, which may be summarized as follows. First, the presence of sediment breaks the symmetry of the flow because of a tendency to self‐stratify. More specifically, this self‐stratification is manifested in terms of a Reynolds‐averaged suspended sediment concentration that declines in the upward normal direction and a Reynolds‐averaged velocity profile with a maximum that is below the channel centerline. Second, this self‐stratification damps the turbulence, particularly near the bottom wall. Two regimes are observed, one in which the flow remains turbulent but the level of turbulence is reduced and another in which the flow relaminarizes in a region near the bottom wall, i.e., bed. Third, the analysis allows the determination of a criterion for the break between these two regimes in terms of an appropriately defined dimensionless settling velocity. The results provide guidance for the improvement of Reynolds‐averaged closures for turbulent flow in regard to stratification effects. Although the analysis reported here is not performed at the scale of large oceanic turbidity currents, which have sufficiently large Reynolds numbers to be inaccessible via DNS at this time, the implication of flow relaminarization is of considerable importance. Even a swift oceanic turbidity current which at some point crosses the threshold into the regime of relaminarization may lose the capacity to reentrain sediment that settles on the bed and thus may quickly die as it loses its driving force.