Jeffrey L. Krolik
Sunil Narasimhan
Dept. of Elec. Eng., Duke Univ., Box 90291, Durham, NC 27708
The ability to measure climatic changes in ocean temperature is fundamentally limited by the presence of mesoscale variability. Because ocean acoustic propagation depends on the range-averaged sound-speed (and hence temperature) profile, long-range acoustic transmissions have been proposed as a means of filtering out mesoscale variability in order to measure a global warming related trend in mean temperature. The Cramer--Rao lower bound (CRLB) on the estimation of a change in the mean depth-dependent temperature profile is presented to determine the highest accuracy which could be achieved by acoustic thermometry. This work extends [A. B. Baggeroer, J. Acoust. Soc. Am. 95, 2850 (A) (1994)] by evaluating the CRLB for different representations of the mean depth-dependent temperature profile perturbation with different levels of a priori knowledge about the mesoscale sound-speed variability. With prior statistical knowledge of the mesoscale variability, the CRLB indicates that accurate measurement of the climate signal may be possible using a general Chebyshev polynomial representation of the mean depth-dependent temperature perturbation. [Work supported by ONR.]