Using a deep source and deep receivers, measurements of the acoustic reflectivity of the ocean bottom at 1 kc/sec were made at an area in the Atlantic Ocean. The data, that is, the direct and the first-order bottom-reflected arrivals, were acquired digitally at sea and have been reduced via a computer program. The program computes the reflection coefficients based on peak pressure as well as the time-integral square of the pressure. The computer analysis also furnishes the travel time of each arrival, horizontal range, and incident angle for each transmission. The total number of data points considered exceeds 4000. Computations based on velocities, densities, and attenuations that have been obtained from an analysis of cores taken in the general area of the acoustic tests have been performed, using a theoretical model of the bottom. This model consists of many absorbing liquid plane parallel layers over one absorbing semiinfinite solid. Comparisons of the experimentally derived reflection coefficient as a function of incident angle are made with those obtained from the model, and the relevance of the complicated layering structure and attenuation of the bottom is discussed.

Bottom-reflection measurements were made in four different geographically smooth, nonsloping bottom areas. Reflection coefficients were computed as a function of angle from data obtained at two single frequencies, two frequency bands, and several pulse lengths. The experimental design and first area results are discussed elsewhere [J. Acoust. Soc. Am. 38, 707–714 (1965)]. Vertical hydrophone arrays were used in three of the areas and it was possible to make comparisons based on hydrophone separation. Average values of reflectivity for data taken at the same angle but for different bottom-reflection points within a small area are observed to differ by as much as a factor of two. Cores taken within the same reflection areas as the observed acoustic results also showed variations in sediment properties. Results obtained at the four locations are presented to show the degree to which it is possible to characterize a large geographic area from measurements taken over a very small area within it.

In a paper presented at the 72nd meeting of the Acoustical Society of America, the author discussed the high variability in the peak amplitude bottom reflection coefficients obtained using pulsed CW signals over an abyssal plain area. These results led to an investigation into the use of a reflected-to-incident energy ratio as a more suitable description of bottom reflectivity. Ratios between the integral-square values of the bottom-reflected and direct-path signal envelopes were computed for 100 msec pulsed CW and 250-msec linear frequency modulated (LFM) signals. The frequencies were 1080 and 3700 Hz for the CW; the LFM bands were centered on these frequencies. Results indicate that the integral-square ratios, which are essentially energy ratios, are as variable as the peak reflection coefficients. The variability, expressed as the interval bounded by the mean plus and minus one sigma, is higher for the 3700-Hz data. For both frequencies and signal types, this variability often exceeds 4 dB over survey areas only a few miles in extent.

No abstract

A theoretical model of the ocean bottom has been developed for studies of bottom reflectivity. The model consists of many absorbing liquid, plane parallel layers over one semiinfinite solid. Using this model, computations of reflection coefficients as a function of incident angle were performed. Measurements of the acoustic reflectivity of an area of Atlantic Ocean bottom, using a deep streamed projector and two deep-fixed receiving hydrophones, were made at a frequency of 1000 cps and for two pulse lengths. The experimentally derived reflection coefficients are analyzed on the basis of pulse length, for each hydrophone, and comparison is made with the predicted coefficients obtained from the theoretical model.