From "STAR" Issue 11 Winter 2001.
A newsletter from OmniSTAR - a member of the Fugro Group.

OMNISTAR used in determining upper seabed structure

In May 2000, personnel from Defence Science and Technology Organisation (HMAS Stirling) and Curtin University's Centre for Marine Science and Technology (CMST) conducted a joint experiment aimed at addressing several objectives. These objectives included:

to explore the possibility of receiving sub-bottom reflections over a large range of grazing angles,
to explore the possibility of receiving head wave signatures to help infer layer depth and compressional velocity,
to explore the acoustical properties and effectiveness of implosive sources such as light globes and purpose-built evacuated spheres over a range of implosion ranges, and
to assess the effectiveness of a single maritime platform, a simple acoustic source such as an airgun, and two receivers to yield useful geoacoustic information about the geology of an experiment site.

This experiment consisted of a single maritime platform towing CMST's airgun and intermittently imploding evacuated spheres and light globes whilst driving away from two fixed receivers (bottom and mid-water mounted) then, after approximately 1.1km range, the airgun was towed at approximately 2.5 knots (roughly 5km/hr) until recording time expired.

Recording package location and vessel position were monitored using an OmniSTAR differential GPS linked directly to a laptop computer with graphical position logging software. This allowed the research team to confirm horizontal separation between the recording package and vessel at all times.

Transmission loss calculations for the mid-water hydrophone data were performed and were compared with propagation model predictions for a geoacoustic profile inferred from the head wave and reflection information. There is reasonable overall agreement at low frequencies (16Hz, 32Hz, and 63Hz) between data and propagation model prediction. In fact, there is a very good agreement at all frequencies within 1.1km range (airgun drifting). However, after 1.1km range (airgun towed at ~2.5 knots), the agreement at higher frequencies (particularly 250 Hz and 500 Hz) is poor. Difference in loss at these frequencies are sometimes as high as 15 dB. This would indicate at least two possibilities:

the geoacoustic model of the experiment site does not describe the area adequately,
there is some physical process affecting these frequencies.

It can only be concluded that sufficient information pertaining to the transverse (shear) wave velocities of the layers would provide adequate agreement between data and model prediction. Preliminary testing of shear velocity and attenuation values has proved successful in obtaining a better fit at all frequencies over the range of the experiment (5.5km).

Sophisticated analysis in the form of geoacoustic inversion is currently being employed to help refine these guesstimates.

References:
Collins, L.B., (1993), "Post-glacial sediments and history, Southern Rottnest Shelf, WA", PhD. Thesis, University of WA.
Article written by Justin Hoffman and John Penrose - story edited for layout purposes.

Figure 1: Experiment location where acoustic transmissions were measured. Depth contours are in metres.

Figure 2: Raising the implosive source housing after imploding a purpose-built evacuated sphere at 40m depth.

Figure 3: Signals from the final track stacked onto the range-reduced travel time plane. The red line outlines the arrival of the shallow interface head wave, and the black line outlines the arrival of the deep interface head wave.