Search

Wave attenuation from two field experiments in the ice‐covered Southern Ocean is examined. Instead of monotonically increasing with shorter waves, the measured apparent attenuation rate peaks at an intermediate wave period. This “rollover” phenomenon has been postulated as the result of wind input and nonlinear energy transfer between wave frequencies. Using WAVEWATCH III®, we first validate the model results with available buoy data, then use the model data to analyze the apparent wave attenuation. With the choice of source parameterizations used in this study, it is shown that rollover of the apparent attenuation exists when wind input and nonlinear transfer are present, independent of the different wave attenuation models used. The period of rollover increases with increasing distance between buoys. Furthermore, the apparent attenuation for shorter waves drops with increasing separation between buoys or increasing wind input. These phenomena are direct consequences of the wind input and nonlinear energy transfer, which offset the damping caused by the intervening ice. A long standing question about the non‐monotonic attenuation rate observed in waves propagating in ice covers is re‐visited. This phenomenon has been discussed for over four decades without a conclusive explanation. Using recent data from two independent and very different field experiments in the Antarctic marginal ice zone, and the most updated global wave model WAVEWATCH III®, the reason for this "rollover" of attenuation is determined. The wind input and nonlinear transfer of wave energy between different wave frequencies offset the damping caused by the ice cover, resulting in the apparent drop of attenuation especially for short waves. Therefore, as the distance between two observation points increase, the decay of high frequency wave energy between the two points decreases. With increasing separation between the two observation points, the wave period for rollover increases and the attenuation rate for all periods drops. When using field data to measure the attenuation rate, care must be taken to consider input from the existing energy sources between the two measuring points. Reduced attenuation at short waves (rollover) is shown with field data to result from wind input and nonlinear transfer between frequenciesThe period from which rollover happens increases with distance between the measuring buoysThe apparent attenuation of short waves drops with increasing distance between buoys and increasing wind field

This paper presents a wave‐in‐ice model calibration study. Data used were collected in the thin ice of the advancing autumn marginal ice zone of the western Arctic Ocean in 2015, where pancake ice was found to be prevalent. Multiple buoys were deployed in seven wave experiments; data from four of these experiments are used in the present study. Wave attenuation coefficients are calculated utilizing wave energy decay between two buoys measuring simultaneously within the ice covered region. Wavenumbers are measured in one of these experiments. Forcing parameters are obtained from simultaneous in‐situ and remote sensing observations, as well as forecast/hindcast models. Cases from three wave experiments are used to calibrate a viscoelastic model for wave attenuation/dispersion in ice cover. The calibration is done by minimizing the difference between modeled and measured complex wavenumber, using a multi‐objective genetic algorithm. The calibrated results are validated using two methods. One is to directly apply the calibrated viscoelastic parameters to one of the wave experiments not used in the calibration and then compare the attenuation from the model with measured data. The other is to use the calibrated viscoelastic model in WAVEWATCH III®over the entire western Beaufort Sea and then compare the wave spectra at two remote sites not used in the calibration. Both validations show reasonable agreement between the model and the measured data. The completed viscoelastic model is believed to be applicable to the fall marginal ice zone dominated by pancake ice. A large data set is analyzed to quantify ocean wave attenuation in the Arctic fall marginal ice zoneThe attenuation data set and multiple wind and ice data sets are used to first calibration and then validate a viscoelastic ice modelCalibrated model parameters depend on wind and ice data used. Results suggest dependence on the ice morphology

A model for wind‐generated surface gravity waves, WAVEWATCH III®, is used to analyze and interpret buoy measurements of wave spectra. The model is applied to a hindcast of a wave event in sea ice in the western Arctic, 11–14 October 2015, for which extensive buoy and ship‐borne measurements were made during a research cruise. The model, which uses a viscoelastic parameterization to represent the impact of sea ice on the waves, is found to have good skill—after calibration of the effective viscosity—for prediction of total energy, but over‐predicts dissipation of high frequency energy by the sea ice. This shortcoming motivates detailed analysis of the apparent dissipation rate. A new inversion method is applied to yield, for each buoy spectrum, the inferred dissipation rate as a function of wave frequency. For 102 of the measured wave spectra, visual observations of the sea ice were available from buoy‐mounted cameras, and ice categories (primarily for varying forms of pancake and frazil ice) are assigned to each based on the photographs. When comparing the inversion‐derived dissipation profiles against the independently derived ice categories, there is remarkable correspondence, with clear sorting of dissipation profiles into groups of similar ice type. These profiles are largely monotonic: they do not exhibit the “roll‐over” that has been found at high frequencies in some previous observational studies. There is strong correspondence between observed ice type and independently estimated frequency‐dependent wave dissipation rateThe dissipation versus frequency profiles do not exhibit the “roll‐over” effect reported in some prior observational and theoretical studiesCases with pancakes in frazil with visual indicators for thicker ice cover have highest dissipation rate, particularly at high frequencies

Wave attenuation coefficients (α, m−1) were calculated from in situ data transmitted by custom wave buoys deployed into the advancing pancake ice region of the Weddell Sea. Data cover a 12 day period as the buoy array was first compressed and then dilated under the influence of a passing low‐pressure system. Attenuation was found to vary over more than 2 orders of magnitude and to be far higher than that observed in broken‐floe marginal ice zones. A clear linear relation between αand ice thickness was demonstrated, using ice thickness from a novel dynamic/thermodynamic model. A simple expression for αin terms of wave period and ice thickness was derived, for application in research and operational models. The variation of αwas further investigated with a two‐layer viscous model, and a linear relation was found between eddy viscosity in the sub‐ice boundary layer and ice thickness. Linear relation found between wave attenuation and pancake ice thicknessLinear relation found between eddy viscosity and ice thickness in viscous modelSimple expression derived to calculate attenuation at any period