Hydropower water level regulation: effects on the ice cover of two Norwegian reservoirs

In regions with freezing lakes, the stability and bearing capacity of lake ice cover are crucial to the safety of communities whose activities revolve around lake ice, and animals crossing the lake’s surface. In hydropower (HP) reservoirs, the development and integrity of the ice cover are influenced by the artificial movement of water, which alters the lake’s flow and thermal structure, and generates rapid and intense changes in water level. Knowledge of ice conditions and of safe limits for water level variations is therefore particularly important in these systems. However, monitoring the ice cover is often logistically challenging and there is still a limited understanding of the extent to which HP strategies for water level regulation can improve or disrupt the stability of the ice sheet, especially in reservoirs with complex bathymetry.

In this study, we investigate the state of the ice cover of two Norwegian HP reservoirs with complex bathymetry and large and frequent variations in water level, where ice monitoring is minimal to absent. We inspected multi-sensor remote sensing data (SAR and optical) over nine winters (2014-2023) to detect the presence of cracks and discontinuities in the ice sheet. We then analyzed water level and meteorological data to identify the primary driver and mechanism of cracking and used simple mechanical and thermal expansion models to interpret the results and isolate the effects of water level variations from those of temperature fluctuations. The satellite data revealed the presence of large cracks in the ice cover of the two reservoirs in each of the nine winters. The cracks consistently appeared in early winter (December/January), propagating from bathymetric obstacles such as sharp-edged coastal protrusions, rocks and islands, and persisted throughout the winter. The results of the mechanical model are consistent with this observation, showing that even a moderate decrease in water level can lead to cracking when an intermediate support (given by the bathymetric obstacle) is present. The analysis of water level and air temperature also supports this crack-formation mechanism, as a predominance of drops in water level is measured prior to the appearance of cracks, while no preferential dynamics in terms of warming or cooling are registered before cracking. These results suggest that the primary mechanism of crack formation in the ice cover of our study sites is the intense stress concentration above bathymetric obstacles encountered by the ice sheet during water level descent. This underlines the importance of investigating the role that modulation strategies of HP operations can play in maintaining or compromising the stability of the ice cover during the critical period for crack formation. We believe that further studies should extend this research to other systems with complex bathymetry and test the effects of environmental constraints on the rate of water level descent in monitored reservoirs.