Field monitoring and numerical modelling of the ice dam at Sundance Rapids

Manitoba Hydro’s Limestone Generating Station, located on the Nelson River in Manitoba, experiences increased winter water levels in the station tailrace and reduced energy production potential due to a large ice dam that forms downstream at Sundance Rapids. Previous work included the development of...

Full description

Bibliographic Details
Main Author: Stafford, Madison
Other Authors: Clark, Shawn (Civil Engineering), Asadzadeh, Masoud (Civil Engineering), Dow, Karen, Malenchak, Jarrod
Format: Master Thesis
Language:English
Published: 2024
Subjects:
Online Access:http://hdl.handle.net/1993/38079
Description
Summary:Manitoba Hydro’s Limestone Generating Station, located on the Nelson River in Manitoba, experiences increased winter water levels in the station tailrace and reduced energy production potential due to a large ice dam that forms downstream at Sundance Rapids. Previous work included the development of a CRISSP2D site model to simulate the complex ice and hydraulic conditions and the establishment of a field monitoring campaign (2019/20) to collect site-specific hydraulic and meteorological data. This research expanded on previous work by performing three years of winter field monitoring (2020/21–2022/23) and assessing the CRISSP2D models’ ability to simulate on-site observations. The goal was to expand the current understanding of the ice dam that forms at Sundance Rapids to aid in the future development of mitigation techniques to reduce the impact on tailwater levels and energy production at the Limestone Generating Station. Analysis of ice dam impacts on tailrace staging confirmed that over the three winters release was driven by thermal conditions. No thermal condition consistently correlated to release, revealing that other factors influence event variability. Quantitative analysis indicated that ice dam strength impacts the magnitude of release events. Qualitative observations found that discharge may impact thermal release by transporting more/less above 0°C water over the rapids. CRISSP2D energy budget component calculations were modified to improve the comparison between simulated and observed. Following modifications, the model was able to simulate ice dam growth successfully. However, it was determined that the existing thermal release mechanism could not capture release as observed on site. Future work should focus on improving the understanding of ice dam release and the factors that impact it (i.e., discharge, ice dam strength) to assist in developing a site-specific release mechanism in CRISSP2D. May 2024 Northern Scientific Training Program