ASA 125th Meeting Ottawa 1993 May

2aAO15. Simulated annealing and genetic algorithm in shear modulus inversion of shallow-water sediments.

Altan Turgut

Dept. of Geosci., Univ. of Tulsa, 600 S. College Ave., Tulsa, OK 74104

The shear modulus of the seabed is one of the most important geoacoustic parameters for modeling acoustic wave propagation in shallow water. A previously developed inversion method [Yamamoto and Torii, Geophys. J. R. Astron. Soc. 85, 413--431 (1986)] enables seabed shear modulus profiles to be determined by inverting measurements of gravity-wave induced sea-bed motion. It is a linearized inversion method that uses local derivative information to improve a starting model in an iterative fashion. Linearized inversions usually depend strongly on the starting model avoiding extensive exploration of model space and tend to be trapped around local minima. In this paper, two new global optimization methods, namely; simulated annealing and genetic algorithm, are introduced to avoid the limitations of the linearized inversion. Performance of these two methods has been tested by using both synthetic and real data and found to be highly efficient for the inversion of shear modulus profile of the seabed.