A comprehensive method to estimate flood levels of rivers subject to ice jams: A case study of the Chaudière River, Québec, Canada

The main difference between an open-water (regular) flood and an ice jam flood is that it is normally the whole river length that is overtopped whereas an ice jam flood is localized to where the jam is located. Comparatively, the regular flood analysis can use the value of the extreme discharge as the main input parameter for a long river section, an ice jam flood needs to account for the probability of jams of various lengths and intensities occurring at specific locations under significantly variable discharges while having several mechanical ice parameters to be considered. Through the case study of the Chaudière River, the methodology presented in this paper demonstrates how to statistically characterize four significant inputs (jam location, jam length, jam properties and river discharge during jam event) into the widely used numerical river water model (HEC-RAS) and how Monte–Carlo simulations are generated to estimate probable ice jam floods along a whole river reach. The purpose of this article is to propose a robust methodology through a case study and asses the sensitivity that historical and mechanical parameters have as to why specific locations along the reach have higher 1:100 AEP ice-induced water levels as to 1:100 AEP open-water levels. HIGHLIGHTS Comparison of 1:100 AEP open-water flood and ice jam flood.; Historical distribution of input parameters.; Using stochastic modelling with HEC-RAS with external control via Python.; Strength of the Monte–Carlo method.