Surface flow and ice rafting velocities during freezing and thawing periods in Nordic rivers

• River ice dynamics differ significantly across studied northern rivers. • Flow velocity in northern rivers varies during the freezing cycle. • Frazil ice was observed on several consecutive days before freezing. • Water movement affects freezing rate more than temperature. • STIV method effective for tracking river ice using time-lapse cameras. In the present climate, almost all high-latitude rivers in northern regions freeze every winter. In the future, however, climate change may cause shifts in freezing, duration and breakup of river ice cover, runoff, and sediment transport. The main aim of the study is to determine changes in surface flow velocity during river ice freezing and melting periods. We also examine the interplay between surface flow velocity and the initiation and progress of freezing and break-up. The study was performed in three Nordic rivers: the Pulmanki (northern Finland), the Koita (eastern Finland), and the Sävar (northern Sweden). They represent different slope and depth conditions and are representative of boreal and sub-arctic channel types. The flow characteristics were measured and remote sensing data for video-based surface velocimetry analyses (STIV) were collected during one season in each river during autumn ice-formation and spring ice break-up. STIV enabled analyses of surface flow velocities, which were compared to direct flow measurements and against freezing (incl. frazil ice, and partial ice cover) and break-up characteristics. Notable variations were observed in the interaction between ice cover and flow velocities among the three rivers. The Pulmanki River showed the most rapid decrease in flow velocities during freezing due to increased friction from the ice cover. Continuous frazil ice formation occurred in the Sävar and Pulmanki Rivers, while the Koita River experienced intermittent episodes. Temperature was key in accelerating freezing during colder days and causing ice break-up during warmer days. During the melting period, the Pulmanki River had the highest flow velocities during melting, while the Koita and Sävar Rivers showed a steady increase as the ice melted. Regression models showed that higher flow velocities reduced ice cover, with air temperature also influencing ice behavior. The STIV method enabled effective detection of flow and ice rafting and proved useful for tracking river ice with affordable time-lapse cameras. These findings provide valuable insights for managing river systems in cold climates.