Shallow-water receptions from the transarctic acoustic propagation experiment

In April 1994 a long-range, low-frequency acoustic propagation experiment took place in the Arctic to test the feasibility of using acoustic methods to observe large-scale thermal variability. Here the characteristics of the received signals at a vertical hydrophone array and a horizontal geophone array located at the edge of the continental shelf in the Lincoln Sea, about 1 Mm from the source, are discussed with respect to the `'forward'' problem of understanding and correctly modeling propagation. Both cw (tonal) and tomographic M-sequence transmissions are analyzed. It is found that phase stability is very good, and that phase changes are almost entirely due to source/receiver motions. Travel times are not quite as stable, but are consistent with the phase observations, showing that phase can be used to measure travel time changes very accurately. Modal decomposition of M-sequence transmissions received by the vertical array shows an arrival structure in rough agreement: with predictions from a coupled-mode calculation. Predictions using an adiabatic approximation appear to work fairly well for modes 1 and 2, but not for higher-order modes which have significant bottom interaction, implying that accurate knowledge of the bathymetry is important in understanding and modeling the signals. Overall, it appears that the forward problem in the water column is known well enough to make an acoustic monitoring program feasible. The acoustic signals were also sensed by geophones placed on the surface of the ice, but it appears that absolute phase and its spatial coherence across the geophone array (although not time stability at any particular geophone) is greatly degraded by the large inhomogeneities present in older ice. (C) 1996 Acoustical Society of America.