On the structure and dynamics of secondary flows over multi-column roughness in turbulent boundary layers

A. S. Sathe 1, W. Anderson 2, M. Calaf 3 and M. G. Giometto 1 · 2025

Read On the structure and dynamics of secondary flows over multi-column roughness in turbulent boundary layers by A. S. Sathe 1, W. Anderson 2, M. Calaf 3 and M. G. Giometto 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-05889.txt.

Introduction

AS Sathe and three co-authors study side flows over rough walls made of many posts, in churning boundary layer flow. Such side flows shape drag, heat flow, and how gases carry scents and dust. Their work came out in twenty twenty-five.

Most past work used neat, even spaced roughness, but real walls are clumped and messy. The team ran large eddy simulations over post clusters with varied density and layout. Shifts in layout flipped the side flow's direction. A key factor was the gap between the edge posts of near clusters, scaled by the channel depth. The authors also track how flow direction shapes fast, snap by snap motion.

Catalog record

CREATORA. S. Sathe 1, W. Anderson 2, M. Calaf 3 and M. G. Giometto 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE93 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-05889

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On the structure and dynamics of secondary flows over multi-column roughness in turbulent boundary layers A. S. Sathe 1, W. Anderson 2, M. Calaf 3 and M. G. Giometto 1 Abstract Secondary flows induced by spanwise heterogeneous surface roughness play a crucial role in determining engineering-relevant metrics such as surface drag, convective heat transfer, and the transport of airborne scalars. While much of the existing literature has focused on idealized configurations with regularly spaced roughness elements, real-world surfaces often feature irregularities, clustering, and topographic complexity for which the secondary flow response remains poorly understood. To address this gap, we investigate multi-column roughness configurations that serve as a regularized analog of roughness clustering. Using large-eddy simulations, we systematically examine the effects of cluster density and local arrangement on secondary flows, and observe that varying the cluster density gives rise to distinct secondary flow polarities. Through a systematic parameter study, we identify the spanwise gap between the edge-most roughness elements of adjacent columns, normalized by the channel half-height ( s a / H s_{a}/H ), as a key geometric factor governing this polarity. In addition to analyzing the time-averaged structure, we investigate how variations in polarity affect the instantaneous dynamics of secondary flows. Here, we find that the regions of high- and low-momentum fluid created by the secondary flows alternate in a chaotic, non-periodic manner over time. Further analysis of the vertical velocity signal shows that variability in vertical momentum transport is a persistent and intrinsic feature of secondary flow dynamics. Taken together, these findings provide a comprehensive picture of how the geometric arrangement of roughness elements governs both the mean structure and temporal behavior of secondary flows. † † affiliation: 1 Department of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY 10027 2 Department of Mechanical Engineering, The University of Texas at Dallas, 800 West Campbell Road, Richardson, TX 75080, USA 3 Department of Mechanical Engineering, University of Utah, Salt Lake City, Utah 84112, USA 1 Introduction Understanding the dynamics of turbulent boundary layers (TBLs) is a long-standing challenge in fluid mechanics, with foundational studies dating back a hundred years ( Prandtl 1925 ; Nikuradse 1933 ; C. F. Colebrook & Taylor 1937 ; Schlichting 1937 ) . TBLs occur over various surfaces, including urban environments ( Oke et al. 2017 ; Oke 1982 ; Meili et al. 2020 ) , wind farms arrays ( Calaf et al. 2010 ; Meyers & Meneveau 2013 ; Cortina et al. 2020 ) , ships ( Grigson 1992 ; Schultz 2000 ) and turbomachinery ( Acharya & Escudier 1986 ) to name a few. When these rough surfaces exhibit spanwise heterogeneity comparable to the boundary layer height, they give rise to roughness-induced secondary flows. Unlike secondary flows of the first kind, which arise from vortex stretching and tilting mechanisms typically induced by curvature or rotation in the mean flow, these are the secondary flows of the second kind which arise from gradients in Reynolds stresses and are maintained through anisotropic turbulent production and dissipation ( Hinze 1967 ; Anderson et al. 2015a ) . These secondary flows, characterized by streamwise vortical structures that redistribute momentum and energy in the spanwise-wall-normal plane, can significantly modify the mean flow field, alter turbulence intensities, and influence quantities of engineering relevance such as surface drag and convective heat transfer. Over the past decade, substantial research efforts have been devoted to advancing our understanding of secondary flows induced by surface roughness heterogeneities. Numerous experimental investigations ( Barros & Christensen 2014 ; Vanderwel & Ganapathisubramani 2015 ; Kevin et al. 2017 ; Medjnoun et al. 2020 ; Wangsawijaya et al. 2020 ; Womack et al. 2022 ) and numerical studies ( Willingham et al. 2014 ; Anderson et al. 2015a ; Yang & Anderson 2017 ; Hwang & Lee 2018 ; Stroh et al. 2020 ; Joshi & Anderson 2022 ; Kaminaris et al. 2023 ; Sathe & Giometto 2024 ) have demonstrated the prevalence and significance of these flows across various roughness configurations. Typically, these secondary flows manifest as pairs of counter-rotating vortical structures whose dimensions scale with the characteristic spanwise spacing of the roughness elements. A distinctive hallmark of these secondary motions is the spanwise undulation of the mean streamwise velocity throughout the flow depth, resulting in clearly delineated low- and high-momentum pathways (commonly termed LMPs and HMPs, respectively) ( Barros & Christensen 2014 ) . Specifically, LMPs are bounded by vortical pairs oriented such that these pathways align with regions exhibiting upward vertical velocities. In contrast, HMPs align with regions of downward vertical velocities. These alternating momentum pathways recur periodically along the spanwise direction, directly reflecting the periodic arrangement of the roughness heterogeneity. The self-sustaining nature of roughness-induced secondary flows is fundamentally tied to spanwise heterogeneities in the Reynolds stress components, which actively generate and maintain streamwise vorticity through turbulence anisotropy and vorticity production mechanisms ( Anderson et al. 2015a ) . Gradients in both Reynolds normal and shear stresses arise due to spatial variations in surface roughness, which modulate the local turbulence intensity and stress distributions across the span. An alternative yet complementary explanation for the maintenance of these secondary motions was proposed by Hinze 1967 , based on the energy budget of turbulent kinetic energy ( tke ). Assuming streamwise homogeneity and negligible transport contributions—assumptions valid for strip-type roughness configurations—the tke budget reduces to a balance between production, dissipation, and advection. In this framework, local imbalances between production ( \mathcal{P} ) and dissipation ( \epsilon ) necessitate advection of tke in the spanwise-wall-normal plane. Regions of excess production draw in turbulence-deficient fluid from the outer layer, inducing a downdraft above high-production zones. This fluid, having acquired elevated turbulence levels due to intense local shear and Reynolds stress production, becomes turbulence-rich and is laterally transported toward regions dominated by dissipation, generating spanwise outflow near the wall. The resulting turbulence-deficient fluid is advected upward, closing the circulation and giving rise to spanwise-periodic roll cells with distinct zones of updraft and downdraft. This cyclic redistribution of turbulence and momentum establishes that secondary flows are sustained by the spanwise modulation of \mathcal{P}-\epsilon imposed by the underlying surface heterogeneity. Analysis of flow over spanwise heterogeneous strip-type roughness has revealed that regions of elevated turbulent production coincide with areas of high roughness, whereas regions of excess dissipation align with low-roughness regions ( Anderson et al. 2015a ) . This spatial disparity in the turbulence budget gives rise to secondary circulations in which LMPs and their associated updrafts emerge over low-roughness regions, while HMPs and corresponding downdrafts align with high-roughness areas. A similar alignment was reported in the experimental study of Barros & Christensen 2014 , where flow over a turbine blade—damaged by the deposition of foreign material—exhibited LMPs over recessed roughness and HMPs over elevated deposits. In contrast, Vanderwel & Ganapathisubramani 2015 , in their experiments involving streamwise-aligned LEGO blocks with varying spanwise spacing (commonly termed “ridge-type” roughness), observed the opposite: LMPs were found over the elevated roughness, while HMPs occupied the recessed regions. This reversal in the alignment of secondary flow features—often referred to as flow polarity—has been a subject of ongoing debate ever since. Direct numerical simulations by Hwang & Lee 2018 of spatially developing turbulent boundary

Read the complete text