Investigation of Anchor Ice Evolution: Numerical Simulations, Field Measurements, and Laboratory Experiments

Frazil ice is generated in the supercooled turbulent water column during river freeze-up, and as it accretes to submerged surfaces, anchor ice forms. Anchor ice formation is known to significantly influence sediment transport, fish habitats, operation of water intakes, hydropower generation, and river hydrodynamics. Important advancements have been made in understanding the environmental conditions leading to anchor ice initiation and release, the impact of anchor ice, and anchor ice structures and properties. However, in-situ measurements of anchor ice evolution are very rare both in the field and laboratory environment. Existing river ice models include an algorithm for simulating anchor ice growth, decay, and release, but there are still many empirical elements and significant uncertainties in the model that influence the accuracy of anchor ice simulations. The motivation for this thesis was to better understand anchor ice evolution under varying hydraulic and meteorological conditions and to improve the accuracy of simulations of anchor ice processes made using numerical models. The anchor ice modeling algorithm was evaluated by comparing to the measurements of two anchor ice events collected during the 2019 freeze-up period in the North Saskatchewan River. The University of Alberta’s River1D model was used to facilitate the analysis. The frazil accretion rate and anchor ice porosity in the equation for simulating anchor ice growth and decay were calibrated to be 1.0 × 10-3 m/s and ~80%, respectively. The calibrated model was able to simulate anchor ice initiation and release times and growth rate for one event accurately, but it failed to predict the timing of another event. This failure was shown to be largely due to the inaccurate simulation of the corresponding supercooling event. Uncertainties in simulations of water temperatures during river cooling and freeze-up periods were then systematically assessed using the University of Alberta’s River1D Ice Process model and field measurements. In particular, the choice of the heat transfer model and the proximity of the weather station to the study reach were investigated. Results showed that the full energy budget model using local weather data was overall the most accurate in simulating water temperatures. The full energy budget model was more accurate than the linear heat transfer model during the freeze-up period when using remote weather data. The linear heat transfer model was insensitive to local versus remote weather data and performed better in predicting the freeze-up starting time. Neither model was able to consistently and accurately simulate the timing and magnitude of observed supercooling events. Field measurements of anchor ice evolution were collected using an underwater imaging system and an artificial substrate. The hydrometeorological data and frazil ice concentrations were also measured. It was found that the temporal evolution of anchor ice occurred in one of three stages: growth, stable, or decay. The variation of the net air-water heat flux was found to be a reliable indicator of the timing of these three stages. Anchor ice growth was mainly through a combination of frazil accretion and in-situ crystal growth, resulting in event-averaged growth rates from 0.48 to 1.77 cm/h. Anchor ice grew solely through frazil accretion when snowfall occurred, with event-averaged growth rates from 1.64 to 3.53 cm/h. The frazil accretion rate was estimated to vary from 2.6 × 10-3 to 6.0 × 10-3 m/s during one event when snow was falling. Anchor ice decay occurred through the thermal thinning of accumulations at average decay rates varying from –0.48 to –2.52 cm/h. Laboratory experiments were conducted in a frazil ice tank to investigate the impacts of steady and varied heat flux conditions on anchor ice evolution. Results showed that the initial, transitional, and final stages occurred in sequence in steady heat flux cases, and an additional heat change stage occurred after increasing the heat flux in the varied heat flux cases. The peak anchor ice growth rate reached a threshold of ~9 cm/h when the air-water heat flux exceeded ~300 W/m2. Anchor ice growth rates during the transitional and final stages increased approximately linearly with the air-water heat flux. The anchor ice growth rate during the heat change stage was 260% larger on average than during the preceding transitional stage, and it was not significantly influenced by the timing or method of varying the heat flux.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.