River Ice Analyses and Roughness Calculations Using Underwater Drones and Photogrammetric Approach

ABSTRACT Freshwater ice forms part of the cryosphere during winter. River ice cover changes the hydraulic resistance of a channel, affecting both water levels and flow velocity distribution. These changes in flow velocity and turbulence depend strongly on the ice characteristics, particularly its hydraulic roughness. However, ice‐covered areas often remain understudied due to challenging conditions and limited access. This study presents an improved approach to assessing hydraulic roughness of river ice using a cost‐efficient underwater drone platform equipped with camera systems. The developed method incorporates a photogrammetric approach, Structure from Motion. A key outcome is a workflow for reconstructing a digital elevation model of the ice underside. Manning's roughness coefficient for the ice was derived from the photogrammetric reconstruction, yielding a value of 0.021 in mid‐winter conditions. Results show that applying Structure from Motion to map the ice underside offers a practical mean to assess hydraulic roughness of river ice, which has broader potential for hydro‐morphodynamic modelling in subarctic rivers under changing environmental conditions. Furthermore, underwater drones allow for more detailed and comprehensive observations of river ice compared to conventional methods.