### ASA 124th Meeting New Orleans 1992 October

## 5aUW4. Statistical analysis on low-frequency sea bottom reverberation
measurement data.

**Li Zhang
**

**
**
*Ctr. for Marine Sci., Univ. of Southern Mississippi, Stennis Space Center,
MS 39529
*

*
*
**Ralph R. Goodman
**

**
**
*Penn State Univ., State College, PA 16804
*

*
*
Six regression formulas are used for fitting the measured sea-bottom
scattering strength SS, which are regarded as functions of grazing angle
(theta) and frequency f. The statistical analysis results are based on the
measurement data from the published papers and technical reports since 1960.
The data measured in the abyssal plain areas are closer to the Lambert's law
(SS=A+20 log sin (theta)) than those measured in the abyssal hills or ridge
regions. But the formula SS=A+B log tan (theta) [A. W. Burstein and J. J.
Keane, J. Acoust. Soc. Am. 36, 1596--1597 (1964)] is better than the Lambert's
law in all three regions. High correlation coefficients of the regressions can
only be found if all the data are from same physiographic province. Otherwise
the correlation coefficients are usually below 0.7, which indicate the current
regression formulas are inadequate to describe the seafloor reverberation. In
the 0.1- to 5-kHz band, the frequency dependence of scattering strength (SS vs
log f) is very poor.