Analysis of Recent Roughness Correlations on Realistic Ice Geometries via Embedded LES

This study uses embedded large eddy simulations of a developing turbulent boundary layer over a heated flat plate with realistic ice roughness to assess local effects on velocity (ΔU+) and thermal (ΔΘ+) log-layer shifts. Validated against experimental data, the numerical setup is used to analyze the mean turbulent flow properties and ultimately assess recent equivalent sand grain roughness ks correlations. Time-averaged velocity and thermal profiles are extracted along the fully-turbulent boundary layer region on top of roughness to access turbulence flow properties, and shifts are computed with respect to the expected smooth behavior. Several established models are taken into account that estimate equivalent sand grain roughness distributions based on geometric properties of the rough surfaces. In order to access a wider interval of k+s, the simulations are conducted at reduced Reynolds number with respect to the reference validation cases, expanding the shifts database for each geometry. The behavior of ΔU+ and ΔΘ+ is observed by employing the different correlations. The results indicate that some existing models capture the trends of velocity and consequently the increase in skin friction fairly well only for high k+s > 400. For lower k+s< 400, these models struggle to correctly capture the expected trends, with except to correlations that incorporate information about surface slope, which suggest improved predictions for the effects of ice. On the other hand, thermal shifts do not exhibit consistent results for any of the investigated models, highlighting that a proper identification of the factors influencing heat transfer enhancement due to ice roughness requires further investigation.