Stanton Scattering Lab Publications

J.D. Warren ⁽¹⁾, T.K. Stanton ⁽¹⁾, M.C. Benfield ⁽²⁾, P.H. Wiebe ⁽¹⁾, D. Chu ⁽¹⁾

⁽¹⁾Woods Hole Oceanographic Institution, Woods Hole MA 02543 USA

email: jwarren@whoi.edu, tstanton@whoi.edu, pwiebe@whoi.edu, dchu@whoi.edu

⁽²⁾Louisiana State University, Coastal Fisheries Institute, 218 Wetland Resources,

Baton Rouge LA 70803 USA

email: mbenfie@lsu.edu

Measurements were made of the acoustic target strengths of siphonophores swimming freely in the ocean. The measurements were made possible by use of a remotely operated vehicle (ROV) on which both acoustics and video camera equipment were mounted. The acoustic transducers and camera were aimed at the same volume of water and data from the two sets of devices were co-registered. The cameras were used to help search for and identify the animals and to direct the path of the ROV; so that once the animals were found they would be in or near the acoustic beams. The data show that these gas-bearing zooplankton have relatively high target strengths due to the presence of the gas inclusion. These data are essential in the use of acoustics for quantitatively surveying zooplankton for two reasons: 1) It suggests that a relatively small number of siphonophores can dominate the echo in an acoustic survey while other species present may dominate the biomass. 2) Siphonophores are very fragile and, as a result, cannot be reliably captured by use of nets. These data can help in making acoustics a quantitative survey tool for these important animals.

1. Introduction

Acoustics provides a means by which marine life such as zooplankton and fish can be rapidly surveyed within a large body of water [Medwin and Clay, 1998]. Knowledge of the target strengths of the animals is critical to the interpretation of acoustic survey data. Target strength information relates the echoes from acoustic surveys to meaningful biological parameters such as length and numerical density. Because of the complexity of the shape and material properties of various marine life, scattering data are required to formulate an accurate model for target strength. Because of the logistical constraints of performing controlled measurements at sea, most scattering measurements have been done in a laboratory. Although laboratory measurements can be highly controlled, there are also artifacts associated with them such as the animal being placed in a shallow tank of water instead of at its natural depth as well as its desired temperature. Thus there is a great need for making controlled measurements of target strength in the natural environment.

One important component of the zooplankton are the physonect siphonophores (Fig. 1). They play an important role in the food chain-- for example, they are predators of copepods, also important animals [Mackie et al., 1987]. Studies of the distributions of siphonophores have been limited, to date, by the inability to sample the animals by the use of nets or pumps. The animals are so fragile that they are destroyed by most mechanical means of capture and thus are greatly undersampled. Further, these colonial organisms are competent swimmers capable of evading slow moving nets. Since the animals have a gas inclusion (pneumatophore), they can be detected by use of acoustics. In fact, they are thought to be potentially significant sources of scattering in the ocean [Barham, 1963]. However, the target strength models of siphonophores, to date, are based upon scattering data collected in a laboratory tank where the animals were in 1 meter of water and subsequently at 1 atm of pressure [Stanton et al., 1998a,b]. Since the gas could, in principle, shrink with increasing depth, the target strengths collected at 1 meter of water may not necessarily correspond to what exists at deeper depths.

Fig. 1. Drawing of a siphonophore. The animal consists mostly of gelatinous tissue with the exception of a gas inclusion at the top (pneumataphore). This gas inclusion can be a significant source of acoustic scattering.

For the reasons given above, acoustical methods are vital to observing distributions of siphonophores and because of the gas inclusions, target strength measurements should be performed in situ as much as possible.

The purpose of this presentation is to describe recent in situ measurements of acoustic target strengths of various siphonophores. The results are discussed in terms of the ability to use sound to quantitatively survey siphonophores.

2. Experiments

The measurements were performed on the RV SEA DIVER using a MAXROVER ROV. The experiments were performed off of Cape Cod, Massachusetts, USA in July, 1998. Abundant layers of siphonophores were observed in this region.

Both video cameras and acoustic transducers (24, 120, and 200 kHz) were mounted onto the front of the ROV and aimed at the same volume of water. The focal point was 1 m in front of the apparatus. The acoustic system was calibrated in the range of depths over which the measurements were performed. Cables from the equipment were tied to the ROV cable and run up into a shipboard laboratory where the video images and acoustic echoes could be observed and recorded. Because of noise transmitted from the ROV cable to the acoustic cables, only the 24 and 120 kHz systems could provide high quality signals. During deployment, the ROV was initially aimed into the current so as to observe marine life swimming or drifting by it. Once an animal of interest was detected with the camera, the operator would then let the ROV drift with the animal while rotating the ROV so as to keep the camera and acoustic beam aimed at the animal. By using this technique, up to one hundred acoustic pings per animal could be recorded.

3. Results

Throughout the several days of measurements, there were many siphonophores observed by the acoustic/video system. The echoes from the siphonophores were generally quite strong (Fig. 2). The echoes were also variable from ping to ping. This variability was due, in part, to the fact that the animals changed location within the beam during the measurements (in spite of efforts by the ROV operator to keep the ROV positioned and aimed so that the animal would remain in the center of the beam). Thus the beam pattern was convolved with the scattering amplitude of the animal.

Fig. 2. Echo from a single siphonophore (24 kHz).

In order to remove the effects of the beam pattern, histograms of the echo amplitudes (once calibrated and adjusted for range from transducer) were compared with classes of curves based on convolution-based echo predictions [Clay, 1983]. These predictions of echo amplitude histograms convolve the scattering response from a target with the beam pattern response. Since the gas inclusion is assumed to be the part of the siphonophore that dominates the scattering and is also much smaller than an acoustic wavelength, the scattering amplitude (echo less beam pattern effects) is assumed to be constant from ping to ping (i.e., no orientation effects). There are no other assumptions on the scattering properties of the animal besides this one. By varying the target strength from prediction to prediction, a class of curves was generated, with one representing the best fit to the data. The target strength associated with the best-fit prediction is then the estimate for the target strength of the animal.

Through this fitting method, the target strength of the siphonophores was approximately -60 dB at 24 kHz. This value is higher than values of target strength measured in the laboratory at higher frequencies. This increase can be attributed to the fact that at these lower frequencies, the resonance frequency of the gas is being approached. For gas bubbles, it is well known that the scattering levels increase as the frequency approaches resonance.

Another important observation is the fact that the scattering levels are consistent with a gas inclusion of about 1 mm, a diameter observed at the surface from animals captured by nets. This strongly suggests that the gas of these animals did not significantly shrink at these deeper depths of the experiment, although the experiments were limited to ROV depths of about 25-30 m.

4. Summary and Conclusions

By use of co-located, co-registered acoustic and video instrumentation mounted on an ROV, target strengths of individual siphonophores were measured. The target strengths were relatively high compared with other zooplankton because of the gas inclusion contained by the siphonophore.

These results are important for two reasons: 1) These in situ measurements confirm earlier reports (from laboratory measurements) of the high target strengths of siphonophores; and 2) The results can be used for interpreting acoustic surveys of siphonophores. Because of the fragile nature of the siphonophores, acoustics is one of the only viable means of quantitatively surveying the animals.

5. Acknowledgements

The authors are grateful to the captain and crew of the RV SEA DIVER for their skilled efforts on this cruise. The authors also gratefully acknowledge Ellen Bailey of the Woods Hole Oceanographic Institution (WHOI) for assistance in the preparation of this manuscript and Dave Gray (WHOI) for assistance on the video images. This is Woods Hole Oceanographic Institution contribution number 9924.

Reference

E.G. Barham, Siphonophores and the deep scattering layer, Science 140, pp. 826-828, (1963).

C.S. Clay, Deconvolution of the fish scattering PDF from the echo PDF for a single transducer sonar, J. Acoust. Soc. Am. 73, pp. 1989-1994, (1983).

G.O. Mackie, P.R. Pugh, and J.E. Purcell, Siphonophore biology, Advances in Marine biology, 24, pp. 98-262, (1987).

H. Medwin, and C.S. Clay, Fundamentals of Acoustical Oceanography, Boston: Academic Press,1998.

T.K. Stanton, D. Chu, P.H. Wiebe, L.V. Martin, and R.L. Eastwood, Sound scattering by several zooplankton groups. I. Experimental determination of dominant scattering mechanisms, J. Acoust. Soc. Am. 103, pp. 225-235, (1998a).

T.K. Stanton, D. Chu, and P.H. Wiebe, Sound scattering by several zooplankton groups. II. Scattering models, J. Acoust. Soc. Am. 103, pp. 236-253, (1998b).