In many Canadian communities, river ice jams pose a significant flood threat each spring. The occurrence of ice jam formation and release events is typically rapid and highly unpredictable;​ thus making their observation and measurement difficult and dangerous. Consequently, knowledge has been limited by the scarcity of reliable data pertaining to these highly dynamic and complicated processes. Furthermore, the capability of forecasting ice jam related floods has been limited to relatively simple conditions.

In this study, numerical models were developed for the purpose of real-time prediction of flood level caused by river ice jam events. Models of varying levels of complexity were built on an open-channel hydrodynamic model:​ River1-D. Given our limited understanding of these phenomena, ice effects were first included into River1-Din a simplified manner, to investigate the extent of ice complexity necessary to successfully simulate flood waves produced by ice jam release events. Full ice effects were then considered in a more complex version of the model, to provide a more comprehensive description of ice dynamics and ice-water interactions during both jam formation and release processes. Applications of the proposed models to hypothetical, experimental, and actual ice jam events provide detailed time-series information of water level, discharge, ice thickness and velocity, thus provided valuable complementary data to the limited available field measurements, and offer a more complete view of river ice jam processes. Field observations were carried out annually during river breakup from 2005 to 2008, for the purpose of obtaining qualitative and quantitative data;​ these have contributed to our understanding of jam processes and have aided the validation of the proposed numerical models.

The work presented in this thesis includes the first numerical assessment of ice effects based on actual ice jam release events;​ the first attempt to numerically simulate how an ice jam breaks through a downstream ice cover;​ and the first attempt to investigate how an ice accumulation responds to applied forces based on experimental data. This research contributes to an advancement of understanding the ice and water components of ice jam processes, and is a step further towards the ultimate goal:​ to be able to accurately forecast water level variations due to ice movements during river breakup.