Load distributions in the ice-structure interaction process in shallow water

Model-scale experiments were performed to explore the process of level ice interacting with a, in model-scale, ten-meter-wide inclined structure in shallow water. During each experiment, an initially intact ice sheet was pushed against the sloping structure, failed against it and accumulated into a grounded rubble pile in front of it. The strength of the model ice used in the experiments was varied. The spatial and temporal ice load distributions were measured at two scales: the structure was divided into ten identical segments along its waterline with the horizontal load on each segment measured, while the local pressure distributions were simultaneously measured with tactile sensors from two of the segments. This paper focuses on the relationships between the loads on different segments and the correlation between global and local loads. The load records of the individual segments during the whole experiment correlated with each other. The Pearson correlation coefficient for the load records of two neighboring segments was about 0.8, and for segments having a distance of 5 m or more, about 0.4. The magnitudes of both the maximum and the mean load on each segment were approximately equal. Even if the segment loads correlated during the whole experiment, the load records of the individual segments showed also transient ice load peaks, which did not correlate. These were used to define local peak load events and related local line loads, defined by dividing the magnitude of a peak load by its width in space. Roughly 2/3 of the local peak load events concentrated on one segment and only about 5% of them covered more than three segments. Narrow local peak load events caused the local line loads of the highest magnitude, with the magnitude and the relative frequency of the events increasing with the ice strength. Narrow events resulted in local line loads up to four times higher than the global line load, which was defined by dividing the total load on the structure by its width. The investigation of the local peak load events also showed that an inclined structure having a width of approximately 5 m (the ice thickness to structure width ratio of 1:100) appeared infinitely wide and gave similar results as a wider structure. Ice pressure measurements showed a line-like pressure distribution at the waterline of the structure with the weakest ice, whereas with stronger ice, zones of high ice pressure were observed at upper parts of the segments.