Experimental investigation of wave-induced bending response of ice floes

Abstract:

Flexural response of ice floes due to waves is one of the many processes involved in the complex wave-ice interactions. On one hand, waves may break up the ice floes through bending processes, which increases the surface area of ice cover, hence accelerating melting. On the other hand, waves are attenuated through bending of ice floes due to the viscous material property of ice, and scattering of waves due to the open leads created by ice fracture. To gain insights into how ice floes respond to waves, well-controlled and carefully-executed laboratory experiments are essential. HYDRALAB+ project: Loads on Structure and Waves in Ice (LS-WICE) attempts to answer some of the fundamental questions regarding 1) fracture of ice due to waves, 2) wave attenuation in broken ice fields, 3) wave-induced motions of ice floes, 4) wave-ice-structure interactions and 5) wave-structure interactions. In the present study, we examine the measurements collected during this experimental campaign to gain further insights into the flexural motions of ice floes. As a result, we find that the elasticity of ice increases the wavelength. We use two multivariate analysis techniques, Proper Orthogonal Decomposition (POD) and Smooth Orthogonal Decomposition (SOD) to identify the dominant flexural modes of ice floes. We observe some weak nonlinearity in the wave-induced flexural response of ice floes and show that both POD and SOD successfully identify the dominant mode shapes (bending responses) that match with the theoretical mode shapes (Meylan and Tomic, 2012). This is useful because the ability to clearly identify dominant mode shapes of ice floes is important for data-driven modeling of wave-ice interactions. In addition, Morlet wavelet time-frequency analysis and Prony’s method are employed to quantify the nonlinearity in the flexural response of ice floes. Results suggest that the response of ice floes to incoming regular waves is virtually linear.