Seasonal ice cover is frequently encountered in rivers located in cold regions that expose to freezing temperatures (Kirillin et al., 2012;Wazney et al., 2019), such as the Nelson River in Canada (Peters et al., 2017), the Mississippi River in America (Ettema, 2002), and the Kanas Lake in China (see Figure 1). Ice cover can have a significant impact on society and the economy. For example, the thawing of the ice cover during the spring season may result in a downstream flood, while the floating ice poses threats to hydroelectric equipment, navigation, and municipal drinking water intakes (Morse & Hicks, 2005;Zare et al., 2016). Due to its practical importance, river ice cover has attracted considerable interest from researchers in the past several decades. Teal et al. (1994) applied point-measurement methods to estimate the vertical profiles of streamwise velocity for the ice-covered channels. Muste et al. (2000) carried out laboratory experiments to determine the effect of the ice cover on the flow fields, turbulence characteristics, and sediment transport rates. Robert and Tran (2012) conducted the laboratory experiment to show that the additional ice sheet and its roughness strongly affected the vertical profiles of the Reynolds shear stresses and the turbulent kinetic energy. Ren et al. (2020) used linearized velocity potential and thin-plate elastic theory to investigate the hydroelastic waves in an ice-covered channel. Other studies of the ice-covered channel flows focused on obtaining the exact solution to the eddy viscosity (Guo et al., 2017); investigating the nonlinear interaction between a floating ice sheet and a water wave (Xia & Shen, 2002); and developing the analytical model of the transverse depth-averaged streamwise velocity distribution (Zhong et al., 2019).The turbulence generated underneath the ice cover will inevitably interact with that generated by the roughness of the channel bed (