The broad-scale acoustic sensors

Introduction to the broad-scale acoustic sensors

The broad-scale acoustic section reviews the use of acoustic techniques for the identification, classification and mapping of benthic habitats. It is intended for readers with an interest or background in the science and technology of acoustic techniques for seabed classification. The utility of acoustic techniques in benthic assessment lies in the ability to provide relatively rapid coverage of large seabed areas, compared to conventional photographic and direct sampling methods. Acoustic techniques should be seen as complementary to aerial and satellite based systems, being of greatest value where factors such as water depth or turbidity limit the scope of optical sensing.

The extensive use of acoustics in marine science and technology reflects the relatively low absorption experienced by acoustic waves in seawater, compared to electromagnetic waves at frequencies of interest. A very substantial literature in marine acoustics is associated with defence and seismic exploration activities, and also with biomass assessment applicable to pelagic fisheries. Acoustical techniques to provide indicators of seabed habitat type are now emerging and are the subject of the CWHM Group's work. Only a small amount of the extensive defence and seismic literature concerned with seabed structures would appear to be of great value, although some techniques from such activities are informing current efforts in habitat assessment. This is because defence and seismic research has normally used relatively low frequencies and involved interactions at considerable depths below the seabed interface, with the exception of some work concerned with the detection of buried munitions. Habitat classification necessarily has a primary focus on the optically discerned water/seabed interface and short distances above and below this boundary. In general this leads to the use of relatively high acoustic frequencies and most examples and discussion in the CWHM Groups work concerns acoustic frequencies at tens of kilohertz or above, essentially the ranges used in fisheries acoustics generally.

Seabed classification is of interest to defence and some engineering tasks. Much defence work has involved comparatively low frequencies, and is not often applicable to benthic assessment. Thus the very extensive literature concerning the development of geoacoustic models for low frequency propagation prediction is not represented here, as the seabed regime of interest for such work extends well below depths of biological importance. Defence concerns for the detection of surface deployed or shallow buried munitions do however overlap to some extent with the issues arising from the classification of benthos, and use relatively high frequencies. A number of extensive defence research programs of this kind have been reported in the open literature, a useful example being provided in the Special Issue on High Frequency Acoustics of the Journal of Oceanic Engineering of January 2001. Simmen et al. (2001) provide an overview of the work covered in this journal. For the purposes of what is presented in the Toolkit, though, direct defence related programs are not reported, as at this time a significant amount of more directly applicable fisheries and environment oriented acoustic systems exist.

Similarly, the need for information about the first metre or so of the seabed, in the context of cable and pipeline burial has led to engineering techniques with some potential for application to broader benthic classification. In Western Australia, Fugro Survey Pty. Ltd. have pioneered a short range, shallow penetration seismic refraction system which provides assessment of the compressional sound speed in the upper few metres of the seabed, a parameter linked to other seabed properties such as compressive strength. At this time, however, this system is unlikely to be competitive with the techniques outlined below and is therefore not further reported here.

Acoustic techniques for seabed classification need to be considered in the context of broader ecological and other considerations. The benthic habitat mapping section deals briefly with some aspects of the benthic environment which have relevance to the use of acoustical techniques.

As with some other techniques, acoustic methods for benthic assessment yield information on surrogate measures of habitat. Acoustic reflection and scattering from the seabed itself and from biota extending above the seabed are central to benthic assessment. Acoustic returns from biota below the seabed surface are not easily distinguished in most acoustic signals. The information presented in this toolkit is compiled at a time when considerable value is seen in mapping seabed habitats such that bottom topography data and acoustic backscatter can be spatially co-located. Such conjoint data sets, informed by periodic towed video information, in particular, are currently seen as providing a workable basis for many seabed habitat requirements. Issues of spatial scales and coverage, of needed and possible spatial resolution, of the choice of classification systems and of survey costs remain as ongoing topics for consideration.

Spatial coverage limitations associated with acoustic systems are particularly significant for, in this respect, the least satisfactory of these systems, those based on single beam echosounders. The relatively low cost of such systems, has nonetheless led to significant use of echosounders in benthic classification. Such single beam systems have been the earliest developed and applied technology in the field. For successful implementation of echo sounding based benthic assessment, a high dynamic range sounder linked with suitable data acquisition and navigation technology is needed. A processing package is called for and several commercial and non-commercial variants exist. Attention needs to be given to several effects arising from varying water depth and to the influence of the host vessel. J. Siwabessy, L.Hamilton and I. Parnum have been lead authors for the single beam techniques and B. Brooke and D.Ryan have contributed additional material. The section on acoustic sediment classification includes some comment on acoustic backscatter from benthic biota, rather than solely from the water – sediment/reef interface itself, a topic of emerging significance. Both of these topics, single beam sounders and acoustic sediment classification, includes discussion on several topics of relevance to the sidescan and multibeam material covered in later chapters.

Sidescan sonar provides extensive spatial coverage and in some cases immediately useful information on bottom type. In general, however, the interpretation of sidescan records is limited in terms of bottom biota and this technology is often seen as providing a preliminary tool to guide a suite of more detailed studies on small areas. Further work using a range of acoustic frequencies and involving texture analysis of sidescan images is warranted to seek maximum value from this technology. The advent of sidescan systems using interferometric techniques to provide linked bathymetry and backscatter information represents, at the time of writing, a developing pathway to the provision of such conjoint information. A. Bickers has been lead author for the sidescan sonar section. The discussion on acoustic system comparisons also contains material on sidescan and interferometric sidescan systems.

The multibeam swath systems now available offer high performance in the provision of topographical information and are also beginning to yield linked backscatter data. The material contained in this toolkit focuses on shallow water examples and applications. One deep water project investigating this potential has however been completed in Australian waters and further deep water swath projects are currently underway. In the presentation of the multibeam and swath systems, A. Gavrilov has been the lead author for this material, which focuses on high frequency, shallow water swath applications. The comparisons of acoustic systems also revisits a number of multibeam related topics.

B. Brooke and D. Ryan are lead authors for the sub-surface acoustic sensing technologies,and also the fine-scale seabed coring. Their contribution points to the opportunity to explore linkages between shallow sub-surface geology and the seabed benthic environment. Their material also highlights the issues of sediment classification and mobility in the assessment of benthic habitat.

P. Kennedy of Fugro Survey Pty Ltd is lead author for the discussion on acoustic system comparisons, which provides a valuable case study involving the comparison of a number of acoustic systems in a selected test area off the Western Australian coast and information on an extensive survey program off the Victorian coast, which was begun as the current document was being completed. This material also contains material on data processing methods applicable to the acoustic systems used. At the end of the discussion on the comparison of the acoustic systems there are a number of important conclusions and comments, drawn from the contributions of the author team, that represent a snapshot of issues associated with the use of acoustic techniques in seabed habitat assessment as of July 2005, in the experience of the authors.

Acoustic system selection

Benthic habitat maps derived from acoustically sensed data are, in general, not biologically or geologically definitive but represent distributions of habitat surrogates. Therefore, there will always be a degree of uncertainty associated with these types of habitat maps, which will depend on systems used, coverage, ground-truthing and the type of habitats being mapped. Furthermore, marine environments are dynamic regions, thus, the distributions of the different habitats will change over time. A sonar signal does not directly identify a seabed type, usually a type must be identified by a ground-truth sample then correlated with the sonar signal. The ability of acoustic systems to identify different habitats/biotopes can only be realised where the seabed or its contents modifies the sonar signal to provide contrast with the background. For instance, an area of seabed that would be considered a biologically different region from another might not be acoustically or topographically different, but needs to be if it is to be distinguished. Notwithstanding these limitations, the capacity of acoustic systems, and the value of the surrogate measures provided from them, are well established so that for many survey requirements the need is to select the appropriate acoustic system, rather than decide if acoustic technology should or should not be adopted. See the section with the comparison between different broad scale acoustic sensors.

Brown et al. (2001) provided a comprehensive set of relevant conclusions at that time, which, in summary were:

A combination of photography, direct sampling and acoustic techniques should be used in seabed assessment
Acoustic techniques should always be used in conjunction with direct sampling methods
Swath acoustic systems out-perform single beam acoustic systems
Acoustic and other techniques are evolving rapidly
As many components of the benthic community as possible should be sampled during benthic assessment
Sediment type and seabed morphology appeared to be the major variables influencing community composition in the areas studied

More recently Kenny et al (2003) have emphasized that selecting the most appropriate acoustic system or systems to utilise for seabed mapping is an important issue in coastal zone management. They proposed that the most important factors to consider when deciding the appropriate acoustic mapping technique are:

Size of area to be mapped
Depth range in the survey area
Object detectability

In addition, it is often the level of resources available that will dictate what acoustic and other mapping techniques are employed. Many potential users in e.g., developing countries and under-funded government departments of developed nations, have limited resources available to them. Hence, a compromise is often made concerning coverage, resolution and the amount of ground truth information obtained. There is thus scope to consider a range of acoustic techniques for benthic habitat assessment, allowing for optimal and sub-optimal methodologies, according to survey purposes and available resources. Ultimately cost is a major factor in the decision of what technique is used, but it is also important when making decisions on the details of surveys to be clear about what type of information is required. The techniques used and the outcomes of mapping projects are typically driven by the types of management decisions that are required although the mapping usually only is a small input into the decision making process.

Comparisons between survey techniques are required to facilitate decisions on the types of technologies that should be employed in a survey. A history of Shallow Survey conferences traces surveys undertaken to compare techniques and equipment for mapping the seabed. The Marmion Marine Park trials discussed in the Marmion case studies compare multibeam systems and traditional and interferometric side scans. Uniquely this survey used the same tracks for each technique, allowing detailed and accurate high resolution comparisons between data sets.

Assuming the use of aerial and satellite imagery is inappropriate (e.g. the water is too turbid or too deep for optics) and there is no requirement for sub-bottom information, there are six particular acoustic remote sensing strategies that could be employed, in increasing order of cost and, as outlined below, largely in benefit to habitat mapping:

Single-beam sonar
Side scan sonar
Side scan and single-beam sonar
Interferometric side scan sonar
Multibeam sonar
Multibeam and side scan sonar

The cheapest and simplest acoustic system to operate is a single beam echo-sounder using one or other form of ground discriminating system (e.g. RoxAnn, QTC View or ECHOplus) or if the raw wave form is recorded a similar type of analysis can be performed independently (see Siwabessy, 2001). Single beam systems are however limited by two issues concerning spatial coverage. One concerns the size of the insonified footprint associated with most single beam sounders (a function of beam width and depth). This may well be sufficiently large to provide inadequate spatial resolution in habitats of interest; thus limiting the object detectability factor listed by Kenny et al. (2003). The second, and often critical spatial limitation factor concerns coverage. Unlike more complex systems, where full area coverage is often a realistic option, single beam surveys commonly do not employ sufficiently close line spacing to approach full coverage, due to issues of vessel time, and thus, costs.

By contrast, side scan and swath systems offer workable 100% coverage for many applications, and can provide textural information, which is useful in identifying morphological features such as sand ripples and bedforms. If quantitative bathymetry is not required, then side scan sonar offers a practical way to map benthic habitat boundaries and geological features. At the time of preparation of this review, it appears that a combination of fine spatial scale bathymetry, perhaps better termed bottom topography, and textural information are of great value in habitat mapping. Thus, for low end users or exploratory surveys, a good strategy is to use a single-beam sounder with ground discrimination processing capability and side scan sonar. Side scan sonar now finds itself sandwiched in cost between the choice of using single beam echosounder or multibeam systems. Brown et al. (2005) report on a recent comparison between single beam and side scan mapping. Although traditional side scans provide little information about bathymetry the high resolution, almost photo realistic images that can be obtained reveal much about seabed types. For a good example of this refer to the Cape Byron Park case study. The wide swath and ease of mobilisation and deployment also maintains the position of side scan in the market.

At the other end of the user spectrum, where more money is available, a multibeam sonar linked with either a side scan system or acoustic backscatter processing of the swath signals themselves represents the optimum approach, such as the work done on the Victorian Marine Parks case studies. Although multibeam systems are more expensive to purchase and run, faster speeds of operation (up to 9 knots), precise positioning and improvements in the quality of backscatter processing, not to mention the confident production of high resolution bathymetric maps are now ensuring that multibeam systems are competing financially for mapping projects. This is especially true where the vessel, crew and logistical costs form a high proportion of the survey cost. Cost efficient mapping of habitats on a broad scale is essential to any successful project. In many cases, the most expensive line item in a project is the vessel and its day rate. Therefore making the most efficient use of ship time is essential in any project. Line turn time, online survey speed and swath width (coverage) are major factors affecting survey efficiency. Traditional survey speeds of 4 knots are required to maintain a side scan sonar fish altitude close to the seafloor, especially in deep water. Deeper water requires more tow cable behind the vessel, which in turn increases line turn time. Accurate positioning of the tow fish requires the use of an Ultra Short Baseline System (USBL), increasing the survey cost even further. With the approaching maturity of Multibeam backscatter acquisition and processing, the prospect of habitat mapping without the use of Side scan sonar is becoming a reality. This removes the requirement for a USBL, winch, topside acquisition system, systems engineer and post processing of tow fish navigation, whilst simultaneously improving safety aspects associated with fish deployment and winch operations. Additional trials should be carried out at high survey speeds of 12 knots or more. This has a significant impact on the efficiency of survey operations. In the Victorian Marine Parks project, the initial survey was carried out using a combination of multibeam to yield bathymetry/topography and side scan to provide backscatter maps. Later surveys in this series are to use multibeam alone, with backscatter maps derived from processing of the swath returns.

Another option is to use one of the interferometric side scan systems, which offer a cheaper alternative to multibeam, but appear to offer reduced performance compared to an optimum swath system. Interferometric side scans have attempted to fill a gap in the market, promising side scan resolution backscatter, with high quality bathymetry over wide swath widths for less cost than multibeam systems. For surveys from relatively small vessels in less than 30 m (and lower cost) in sheltered waters, this equation is likely to be true. In surveys where logistical costs are higher however and more expensive and competent motion sensors are required to cope with worse conditions, savings in costs by using an interferometric side scan over a multibeam system may be negligible in comparison with total project cost.

Figure 1 represents one aspect of the comparison between the acoustic systems discussed, noting that all systems represented apart from the single beam example will often lend themselves to full coverage operation.

Outline of object detection ability vs cost of operation for several acoustic systems

Figure 1. Outline of object detection ability vs cost of operation for several acoustic systems

The relationship between the immediate sub-surface constitution of the seabed, including in-benthos organisms, and the visually observed surface and epibenthos is of continuing interest. In this regard, the acoustic systems described in shallow seismic profiling section are of importance. The information that sub-bottom sensing can provide on near surface geological structures is often significantly related to the recent history of the seabed. The emergence of instrumentation providing high levels of depth resolution may also facilitate routine investigation of structure within the top metre or less of sedimentary seabeds.