Author(s): Document owner: Reviewed by: Workgroup: MESH action:

Transcription

1 Title: Author(s): Document owner: Reviewed by: Workgroup: MESH action: Version: Case Study: Optimal detection and identification of biogenic structures using side scan sonar Bill Meadows (CEFAS, Lowestoft), Dave Limpenny and Jacqueline Eggleton (CEFAS, Burnham Laboratory) and Bob Foster-Smith, Envision Mapping, Ltd. Bob Foster-Smith Clare Fitzsimmons, Envision Mapping, Ltd. Action 5.2: Case histories on the applications of habitat mapping Action 3.1 Testing through new surveys 0.1 DRAFT Date published: 30/10/2006 File name: Language: Number of pages: 12 ENV SC01 SSS ID Biogenic Structures.doc English Summary: Reference/citation: Biogenic reefs shelter fauna which is distinct from other biotopes, and allow species to become established in predominantly sedimentary areas where they would not otherwise be found. Therefore, S. spinulosa has been identified as a priority habitat under the Biodiversity Action Plan (BAP). Consequently, more rigorous techniques are required for monitoring such systems. Side scan sonars are generally considered to give the highest definition of seabed features amongst all the standard acoustic remote sensing techniques used for habitat mapping, and have often been recommended as the system of choice for the detection of reef-like features. Part funded by MESH and the Aggregates Levy Sustainability Fund, this case study describes a novel survey which attempted to adopt a structured approach to side scan deployment over areas where reef was known to occur, in order to draw up guidance for increasing the likelihood of detection of reef-like features, and describe limitations to the interpretation of such images. ISBN or full reference TBC

2 Keywords: Side scan Sonar, Sub-Littoral, Mapping, Reefs, Biogenic, Identification, Sabellaria spinulosa Bookmarks: Related information: Links to related pages of websites/related documents/reference numbers of final products MESH metadata: MESH metadata template: MESH

3 Case Study: Optimal detection and identification of biogenic structures using side scan sonar 1. Introduction Biogenic reefs shelter fauna which is distinct from other biotopes, and allow species to become established in predominantly sedimentary areas where they would not otherwise be found. Therefore, S. spinulosa, a reef builder of this type, has been identified as a priority habitat under the Biodiversity Action Plan (BAP). Consequently, more rigorous techniques are required for monitoring such systems. Side scan sonars are generally considered to give the highest definition of seabed features amongst all the standard acoustic remote sensing techniques used for habitat mapping. As such, they have often been recommended as the system of choice for the detection of reef-like features. There are many reports of the detection of Sabellaria spinulosa reefs through the inspection of side scan images, and the identification of unusual textures thought to be caused by the reefs. However, detection of reefs in this way has been largely serendipitous and no previous studies are known to have adopted a structured and methodical approach to testing the success of side scan for this purpose. Side scan performance depends not only on the characteristics of the system itself (e.g., analogue/digital; frequency), deployment (e.g., swath width, tow speed, fish height above seabed, direction of tow) as well as environmental conditions (e.g., swell may result in tugging the fish), the image is also affected by processing (e.g., mosaicing may result in degrading the resolution). Sidescan has been particularly successfully used for mapping biological features where the features are physically distinct and/or obvious, including for a number of biogenic reef features, for example Modiolus reef areas. This study attempted to adopt a structured approach to side scan deployment over areas where reef was known to occur, or where reef had been recorded in the past, in order to draw up guidance for increasing the likelihood of detection of reef-like features and describe the limitations to the interpretation of features seen on the images in terms of reef structures Goals of Mapping Work to date with side scan sonar systems indicates that the main advantage of sidescan is that it can provide high resolution images (approx. 30cm pixel size) with good positional accuracy if it is combined with pitch, heave and roll sensors as well as differential GPS (dgps). However, the main problems with this approach have been; (a) that this technology has yet to obtain clear and distinctive images of Sabellaria reefs, and (b) if reefs are variably developed then it might be difficult to detect the full range of reefs against a background of other habitats. Therefore, the purpose of this work was to determine the optimal use and deployment of side scan sonar systems for the detection of small topographic and texture features associated with biogenic reef-like structures Pilot Sites There are only a few areas of known Sabellaria spinulosa reef in UK waters (and very few in other European waters). The primary site used for testing was an aggregate extraction site, known as site 107, off the Lincolnshire coast. Selected areas within this site were chosen on the basis of video tows that showed there to be well developed Sabellaria reef, areas of non-reef sand and other sand wave features, which were ideal for the purposes of the study. Other sites Author: Bob Foster-Smith File: ENV CS01 SSS ID Biogenic Structures Page 3 of 12 Last saved: 14/11/ :26 Partner: Envision Mapping

4 were also used. An area of Saturn reef was the secondary focus area for this work. Located in the Southern North Sea, Saturn Reef is partially composed of sandbank, with the reef itself rising well above the surface of the seabed, this area contains areas of dense clumps of reef, believed to be well developed Sabellaria spinulosa. However, detection was on the basis of unusual texture on side scan images in 2002, and in recent years this area had been resurveyed and well developed reef was not found. Video tows during the current surveys also confirmed that reef was not present, but assessed the suitability of the site for Sabellaria habitat, and provided a control. Lastly, another aggregate extraction site near Hastings was used having established the presence of well developed Sabellaria reef (16-18 th May 2006 survey) Methods Summary A number of different comparisons between systems and deployment strategies were tested during a cruise on RV Endeavour between the dates 13 th July and 4 th August Three side scan sonar systems were available for use: Benthos SIS kHz chirp with ISIS acquisition, Benthos SIS 1624 dual frequency 100/400kHz chirp with ISIS acquisition and Geoacoustics ss941 switchable 100/400kHz with Coda acquisition. The data were collected in XTF format and processed using ISIS/Delphmap post processing suite. Classification of local habitats, including areas of suspected Sabellaria reef, had previously been accomplished through the analysis of AGDS and appropriate ground truthing using video and grab sampling. This work was used to guide the location of suitable sites, upon which to test the capabilities of the side scan sonar units, in this context. Once such areas had been positively identified, expert interpretation of the resulting side scan sonar images was required to test the efficacy of the system for identification of biogenic reef structures. Comparisons between the systems and deployment methods could then be made, and the optimal use and deployment of side scan sonar systems for the detection of small topographic and texture features associated with biogenic reef-like structures could be determined. This case study report primarily focuses on the 107 aggregate extraction site off the Lincolnshire coast, but draws conclusions based on all three sites (107, Saturn reef and the Hastings aggregate extraction site). 107 had varied seabed features, for the purposes of broad scale comparison, and a very distinctive feature (nicknamed the gorilla s head) which was used for many of the direct comparisons between images. The experimental design at the 107 site was as follows:- 1. Comparison between analogue and digital systems at similar low frequencies (100 or 200kHz) and high frequencies (400kHz); 2. Using the best system, comparing east/west and north south lines run over reefs; 3. Running the dual frequency system at different ranges (150m and 50m) so, comparing both frequency and range; 4. Running the dual frequency system on short range (50m) at a height of 12 and 6 metres; 5. Running the dual frequency system at 100m range and 12m height at 6 knots and 3 knots.

5 2. The characteristics of the habitat type The ross worm, Sabellaria spinulosa Leuckart 1849, is a sedentary, epifaunal polychaete that builds rigid tubes from sand or shell fragments. It is a suspension feeder that is generally found individually but can be gregarious in favourable conditions, and colonies consisting of fused sand-tubes may form thin crusts or extensive reefs. The reefs are solid but fragile structures, which can be up to several metres across and raised above the sea bed by up to 30cm. The video showed that the reefs in the surveyed area of 107 to be wave-like structures about 15-30cm tall with non-sabellaria in the wave troughs. The long axis of the reefs ran east/west. Sediment profile information further showed the Sabellaria tubes to originate from underlying sand waves and not a cobble base. Significantly, the reefs can persist for many years and as such, they provide a biogenic habitat that allows many other associated species, including epibenthos and crevice fauna, to become established. As such, the fauna is distinct from other biotopes and species can become established in predominantly sedimentary areas where they would not otherwise be found. In this guise, therefore, S. spinulosa has been identified as a priority habitat under the Biodiversity Action Plan (BAP), the UK's initiative to maintain and enhance biodiversity. In addition, the reefbuilding sabellariids (S. spinulosa and S. alveolata, Linne) have formed a theme in the LIFE programme promoting management of the candidate Special Areas of Conservation (csacs) Classification On the basis of the definition of biotopes, or biotope types, as the spatial components of an ecosystem characterized by specific ecological, unique and more or less constant environmental conditions, S. spinulosa habitats have been classified according to the biotopes listed in Table 1, below. Both biogenic reef biotopes and non-reef biotopes of Sabellaria have been included here for the purposes of comparison. Table 1: Sabellaria spinulosa biotopes of five different classification systems. Classification Code Biotope(s) Description Britain / Ireland (MNCR BioMar 97.06) CMX.SspiM x MIR.SedK.S abkr S. spinulosa and Polydora spp. On stable circalittoral mixed sediment S. spinulosa with kelp and red seaweeds on sand-influenced infralittoral rock The tube-building polychaete S. spinulosa at high abundances on mixed sediment, with Polydora spp. tubes attached. Infauna comprise typical sublittoral polychaete species, together with the bivalves Albra albra and Nucula nitidosa. Epifauna comprise calcareous tubeworms, pycnogonids, hermit crabs and amphipods. This biotope is considered a biogenic reef since Sabellaria performs an important stabilising function on the substratum. S. spinulosa, sediment-tolerant red seaweeds and occasional Laminaria hyperborea characterise this biotope. Some of the richer examples of this biotope (e.g. Luce Bay) also have a rich fauna of ascidians, sponges, hydroids and bryozoans. A similar biotope is also found in the circalittoral zone, where it lacks the algal component (MCR.Sspi). NB This biotope is

6 MCR. CSab.Sspi MCR.As MolPol.Sab S. spinulosa crusts on silty turbid circalittoral rock Dense ascidians, bryozoans and hydroids on a crust of S. spinulosa on tide-swept circalittoral rock Biodiversity Action Plan Priority Habitats (UK) Europe (EUNIS Nov. 1999) A3.6/B- MCR.Csab A4.4/B-CMX.SspiMx A3.234/B-MIR. SedK.SabKR A3.631/B-MCR. CSab.Sspi A3.6721/B-MCR.As. MolPol.Sab Wadden Sea (1996) S. spinulosa reefs S. spinulosa communities on circalittoral rock S. spinulosa and Polydora spp. on stable circalittoral mixed sediment S. spinulosa with kelp and red seaweeds on sand-influenced infralittoral rock S. spinulosa crusts on silty turbid circalittoral rock Dense ascidians, bryozoans and hydroids on a crust of S. spinulosa on tideswept circalittoral rock Sublittoral Sabellaria reef France (ZNIEFF-MER) III III Faciès à S. spinulosa Faciès d épifaune à S. spinulosa not considered a biogenic reef. Bedrock in moderately exposed, slightly tide-swept conditions with high turbidity with an almost entire crust of S. spinulosa tubes; few other species present. Ciona celata, Alcyonium digitatum and Hypoglossum hypoglossoides present in NE England, very extensive Mytilus edulis in South Wales (Gower). The fauna attached to the Sabellaria crust in many cases seem to reflect the biotopes on nearby rock. NB This biotope is not considered a biogenic reef since many of the associated species are capable of living on the rock irrespective of the presence of S. spinulosa Tide-swept rock in areas with high levels of suspended sand with a S. spinulosa crust which supports a wide variety of other species. A dense carpet of ascidians Molgula manhattensis, Polycarpa spp. and Polyclinum aurantium, a turf of bryozoans (Cellaria sinuosa, Bugula plumosa and Flustra foliacea) and sponges such as Scypha ciliata and Polymastia spp., bryozoans Alcyonidium diaphanum and Scrupocellaria spp. and Antedon bifida may also be present. In some cases this biotope occurs adjacent to MolPol although in deeper water and more tide-swept (scour/turbulent) conditions. NB This biotope is not considered a biogenic reef. See MNCR MCR.CSab See MNCR CMX.SspiMx See MNCR MIR.SedK.SabKR See MNCR MCR.CSab.Sspi See MNCR MCR.As.MolPol.Sab

7 2.2. Relationships (i.e. substrate to biota) Thicker, and probably more permanent, crusts or reefs of S. spinulosa seem to have a considerable influence on the benthic community structure, and have been reported to contain a more diverse fauna than nearby areas. The UK National Rivers Authority found that sites in the Wash associated with S. spinulosa had more than twice as many species and almost three times as many individuals (excluding the Sabellaria themselves) as sites with low, or no, S. spinulosa. Numerous species have been reported in association with S. spinulosa, although virtually all are found widely in other communities; for instance, S. spinulosa communities with attached Polydora tubes, and with an infauna of typical sublittoral polychaete species, as well as the bivalves Abra alba and Nucula nitidosa, and an epifauna including tubeworms, pycnogonids, hermit crabs and amphipods. Previous studies carried out in the Wash by Envision Mapping, also reported associated anemones, as well as high densities of shrimp like organisms, probably mysids, which could be seen immediately above the reef on video footage. The commercially valuable pink shrimp, Pandalus montagui, seems to have a particularly strong association with S. spinulosa reefs to the extent that fishermen pursuing Pandalus have been reported to use small trawls to search for lumps of S. spinulosa which they regard as an indication of good fishing grounds. 3. Summary 3.1. Discussion of results from the techniques (area 107) Comparison of Analogue and Digital Systems There was very little difference between the images using the analogue and digital systems. The analogue system gave a more uniform image across the swath than either of the digital systems and was less responsive to bulk backscatter (an acoustic property more dependant upon sediment properties than seabed relief features). For ease of use, the remaining tests used the digital systems only. Geoswath SS kHz, Benthos SIS kHz, Figure 1: The reefs appeared to produce a ripple-like texture on the images with an east/west long axis. This is reported to be typical of Sabellaria reefs.

8 Perhaps not unexpectedly, the east/west images differed from the north/south images. The ripple-like features described above were much clearer in the east/west runs than the north/south runs. This is expected assuming the orientation of the reefs to be roughly east/west. However, the features must be interpreted with caution since there are signs of tearing along the direction of tow. This can give the impression that features along the axis of travel are less distinct than those perpendicular to it. East/west 400kHz, 6 knots, 50m swath North/south 400kHz, 6 knots, 50m swath Figure 2: Differences in east/west and north/south images. Characteristic Sabellaria ripple-like features are much clearer in the east/west runs Comparison of High and Low Frequencies Generally, side scan sonar systems working with high frequencies have a much lower depth range than low frequencies. Theoretically, high frequencies should give more definition, but the results from these surveys show little difference in resolution, whilst the high frequency images to be much more susceptible to tearing than the low frequency. In addition, the low frequency appears more responsive to bulk backscatter. See Figure 3, below.

9 Figure 3: Differences in side scan frequency employed. The high frequency image (right) appears much more susceptible to tearing than the low frequency image (left) Varying Tow Heights The lower tow height (6m) gave more distinct features on the images for the reefs than at 12m. The difference was more of contrast between the ridge crests and troughs rather than revealing features not recorded at 12m. 6m height (100kHz, 6 knots, 50m swath) 12m height (100kHz, 6 knots, 50m swath) Figure 4: Differences in tow heights illustrated. On the left is 6m height (100kHz, 6 knots, 50m swath), compared to 12m height (100kHz, 6 knots, 50m swath) on the right.

10 Varying Tow Speeds Little improvement was made from running the system at a slower speed (3 knots), though tearing is increasingly marked with the combination of higher frequency and higher speed. 100kHz, 3 knots 400kHz, 3 knots 100kHz, 6 knots 400kHz, 6 knots Tearing Figure 5: The benthos SIS 1624 with 100kHz (left) and 400kHz (right) run at 3 knots (top) and 6 knots (bottom) Comparison with Results from Other Sites Saturn Reef The video tows at Saturn reef show sand and gravel small wave features without any Sabellaria tubes being obvious. Nevertheless, the texture of the images was similar to the images from reefs at 107, although perhaps slightly less distinctive. This highlights the risk of assuming features seen on the images to be associated with reef-like features on the basis of a signature catalogue without adequate ground truthing. Clearly, other ground types can result in similar textures.

11 Hastings The images obtained from Hastings also showed a similar texture to that from the Sabellaria reefs at 107 and the non-reef ground at Saturn reef. However, the reefs at the Hastings site are different in structure from those at 107. The Hastings reefs were not purely Sabellaria, but also included cobble, shell and often large quantities of Crepidula. Thus, again the texture of the images cannot be assumed to be caused by Sabellaria. 4. Recommendations Although performance may be enhanced through use of high frequency side scan (400kHz) in theory, the benefits are by no means clear, and the drawbacks of increased tearing and reduced swath width considerable (Figure 3). The benefit of the response of the lower frequency ( kHz) to bulk backscatter may not be relevant to reef detection in this case, but might be more significant in other scenarios. Again, the benefits of slower speed and reduced height of the fish are marginal and must be set against the drawbacks of increased survey time and danger to the fish respectively (Figure 4 and Figure 5). The risk of attributing textural features on the images to supposed seabed features is very real at the limits of topographic resolution. Clearly, the general guidance that lines should be run at right angles to each other is important, given the variation in the east-west and north-south images (Figure 2). However, in reality many surveys may involve very long tows (e.g., along a pipeline or cable route) and running lines across the route would increase survey costs considerably. 5. Conclusion In order to monitor S. spinulosa reefs, as opposed to annually forming thin crusts, it is important to be able to distinguish and detect their whereabouts and extent. If they are very dynamic, as seems to be the case, a technique for seeing the reef would be invaluable in mapping the way reefs fluctuate. Such detailed maps could be used to help evaluate the significance of changes in the reef at any precise point, whether or not the change is due to fine scale dynamics or to some broader scale trend, for example. However, these surveys conclude that many biogenic reefs are at the limits of resolution of side scan technology, especially if the reefs are to be distinguished from a heterogeneous sea floor. Though it achieves the required resolution, sidescan failed to distinguish clearly between S. spinulosa reef and nearby patchy hard bottoms. Use of acoustic shadows for calculation of the heights of reefs must be regarded as doubtful at this stage.

12 6. Acknowledgements A series of surveys underpinning this case study were part funded by MESH and the Aggregates Levy Sustainability Fund.