Acoustic detection of undersea objects is difficult due to the uncertain environment and even more difficult when the objects are buried in the seabed. First, sediments generate high backscattering noise due to heterogeneous scatters within the sediments, clouding the object.

Second, the acoustic wave attenuation in sediments is much higher than in water. Acoustic shadows make the buried targets absent in the sonar images due to diffraction around the target, transmission through the target, and relatively high acoustic noise due to backscattering from sediments surrounding the target. Classification of buried targets is also more difficult since there are no shadows, and the images do not contain much information about target shape since scattering from oblique target surfaces is not detectable.

The Navy’s Comprehensive Acoustic Simulation System (CASS) was used to investigate such an effect. CASS is the Navy’s standard model for acoustic and sonar analysis. It incorporates the Gaussian Ray Bundle (GRAB) eigenray modes to predict range-dependent acoustic propagation in the 150 Hz to 100 kHz frequency band. CASS calculates the reverberation in nested do loops, seven deep. Reverberation is a function of time, and the inner loop collected all the reverberation contributions over user-requested sampling times. There are two loops on eigenray paths, one for the paths connecting the transmitter to the scattering cell, and the other for the paths connecting the scattering cell to the receiver. Since the reverberation is calculated in the time domain and there may be contributions in the same time bin from different ranges, the next loop increments the range. CASS combines the five possible eigenray paths at each range step and decides if the ray paths contribute to the reverberation time bin.

Test rays are sorted into families of comparable numbers of turning points and boundary interactions. Ray properties are then power averaged for each ray family to produce a representative eigenray of that family. Target echo level and reverberation level are computed separately, and subtracted to get the signal-noise ratio in the absence of additive ambient noise — noise level is typically power summed with the reverberation level for total interference.

CASS is integrated with the sonar parameters, sound speed profile (SSP), bottom type, bathymetry, and the scattering characteristics. The model output of the seafloor reverberation is used to represent the model generated sonar imagery (MGSI). An increase of the volume scattering strength in the lower water column reflects the presence of the suspended sediment layer. Clearly, the object is visible in the reverberation imagery. Since MGSI is a replica of the sonar image, the effect of suspended sediment on acoustic detection may be studied using MGSI.

Suspended sediment increases the volume scattering strength. To simulate its effect, the volume scattering strength is increased by an increment of 5 dB from the value of -65 dB below 78 m depth while keeping the volume scattering strength constant (-95 dB) above 78 m depth, and CASS is integrated with increasing the value of the volume scattering strength to investigate the effect of the suspended sediment on the object detection.

As the volume scattering strength of the sediment layer (below 78 m depth) increases to a value of -30 dB, the object becomes nearly undetectable. The CASS modeling continues with the increase of a smaller increment of 1 dB. The mine-like object is completely obscured by the suspended sediment layer at a value of -22 dB (below 78 m depth), which is taken as the threshold. This threshold (-22 dB) is large compared to existing observational data.

While CASS is useful, several shortfalls remain. First, the process by which the environment and the object are modeled is cumbersome. Second, the appropriate volume scattering strength for the buried object should be given. A thorough study is suggested on the relationship between the suspended sediment layer density and type (e.g., sand, silt, or clay), particle density in the layer, associated volume scattering strength and attenuation, and changes in the sound speed profile.

This work was done by Peter C. Chu and Michael Cornelius of the Naval Postgraduate School, and Mel Wagstaff of the Naval Oceanographic Office at NASA’s Stennis Space Center for the Office of Naval Research. ONR-0027