An experimental study of the coherent under-ice reflectivity of sound in the Greenland Sea Marginal Ice Zone (MIZ)

Abstract

The coherent component of acoustic under-ice reflectivity was investigated as a function of frequency and grazing angle in the MIZ. Measured reflectivity was compared with scattering theory predictions to test the hypothesis of ridge scattering dominance in the MIZ. Explosive source acoustic signals were received on a vertical array, during the Marginal Ice Zone Experiment (MIZEX 84), and deconvolved to separate the direct and ice-reflected arrivals. Ice-reflected signals received on individual hydrophones at different ranges and depths, and on different days, were aligned in time and coherently ensemble averaged within a moving 10-deg grazing angle window. The resulting reflectivity was found to decrease with increasing frequency from 64–256 Hz and with increasing grazing angle from 12–35 deg. Comparison was made with predictions from smooth elastic plate theory, perturbation theory for a rough pressure release surface, perturbation theory for a rough elastic surface, Twersky theory for hemispherical bosses on a plane, and Twersky theory for infinite, semielliptical cylinders on a plane. The Twersky theory for semielliptical cylinders gave a significantly better fit to the data than the other theories. This suggests that, although the under-ice topography may be a good deal more complicated, scattering may, nevertheless, be dominated by remnants of elongated pressure ridge keels. Remote sensing data, taken during the experiment, showed that the ice cover was primarily composed of small floes (< 200 m in diameter). Laser ice-surface height measurements were analyzed to obtain the pressure-ridge sail-height distribution. The Twersky theory modeling gave predictions within 1 dB of the measured mean reflectivity when a ridge keel-to-sail ratio of 6.5 was assumed.