Abstract

Despite a recent resurgence of observational studies attempting to quantify the ice-induced attenuation of ocean waves in polar oceans, the physical processes governing this wave attenuation phenomenon are still poorly understood. Most analyses have attempted to relate the spatial rate of wave attenuation to wave frequency, but have not considered how this relationship depends on ice, wave and atmospheric conditions. An in-depth analysis of the wave-buoy data collected during the 2017 PIPERS programme in the Ross Sea is conducted. Standard techniques are used to estimate the spatial rate of wave attenuation $\alpha$ and the influence of a number of potential physical drivers on its dependence on wave period $T$ is investigated. A power-law is shown to consistently describe the $α(T)$ relationship, in line with other recent analyses. The two parameters describing this relationship are found to depend significantly on sea ice concentration, mean wave period and wind direction, however. Looking at cross-correlations between these physical drivers, three regimes of ice-induced wave attenuation are identified, which characterise different ice, wave and wind conditions, and very possibly different processes causing this observed attenuation. This analysis suggests that parametrisations of ice-induced wave decay in spectral wave models should be piecewise, so as to include their dependence on local ice, wave and wind conditions.