The dynamics of ice-water interfaces: experimental and numerical investigation of ice scallops

Anyidoho, Victor (2023) The dynamics of ice-water interfaces: experimental and numerical investigation of ice scallops. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (3MB)

Abstract

Floating glacial ice features, such as icebergs and ice islands may develop naturally textured ice surfaces due to melt processes, which can lead to the formation of local surface undulations commonly referred to as scallops. The mechanism for the formation of ice scallops was investigated using experimental and numerical techniques to help understand how these features are formed in nature. The variables investigated include temperature, scallop wavelength, and scallop geometry. A series of tests were conducted in a recirculation flume under specific experimental conditions at Reynolds numbers 23265, 30147, and 32768. Each test experiment was run for 10 minutes and observed for scallop formation. The experimental time frame was extended to 20 minutes to see if any additional scallops would be formed. Numerical simulation was conducted to complement the experimental results using commercial software. In both experimental and numerical investigations, the results showed a dominance of a visible recirculation region at the leading edge of the ice. The experimental results showed an increased ice melt rate when the approach fluid velocity was increased. Scallops were formed due to turbulence flow structures and spontaneous separation and reattachment of flow at the leading edge, giving rise to non-uniform melting at the ice-water interfaces. Three main phases of scallop development were identified: an initial flat ice geometry, a fully developed scallop, and an adjusting or evolving scallop. Due to the three-dimensional nature of the flow, portions of the water were directed to the lateral side of the ice, which led to a differential melt. The experimental results also showed a significant melting at the ice front or the leading edge of the ice. This is partly attributed to stagnation and pressure fluctuation within the recirculation bubble. The basal melt profile of the ice was measured using a contour gauge after each experimental run. The dynamic interaction of the water and ice interface resulted in shearing on the ice surface. The average basal melting was estimated at 40 mm relative to the initial origin of the ice. The frontal melt varied from 30 mm to 50 mm when the Reynolds number was increased. The surface of the ice closest to the middle of the channel experienced a higher melt rate than other bottom surfaces of the scalloped ice since the maximum velocity occurs at the center of the channel. Scallops were observed to be formed due to fluid recirculation within the scalloped region, resulting in the evolving scallop geometry. A test to investigate the effect of Reynolds number on scallop wavelength showed a wavelength of 266 mm for Re = 32768, while no wavelength was observed for Re = 23265 and 30147. From the results, a partial feature was observed at the tail end of the ice for Re = 30147 but could not be classified as a full scallop. The measured temperature data indicated increased melting with an increasing Reynolds number. The temperature recorded for each thermocouple channel is observed to reduce from the leading edge to the tail end of the ice. Data from the numerical simulation showed fluid separation at the leading edge of the ice and subsequent reattachment of the fluid, which corroborates the non-uniform melting of the ice observed experimentally.

Actions (login required)

View Item