© 2006 International Council for the Exploration of the Sea
The REBENT monitoring network, a spatially integrated, acoustic approach to surveying nearshore macrobenthic habitats: application to the Bay of Concarneau (South Brittany, France)
Ifremer, Département Dynamiques de l'environnement côtier, Technopole de Brest-Iroise BP 70, 29280 Plouzané, France
*Correspondence to A. Ehrhold: tel: +33 298 224319; fax: +33 298 224548. e-mail: axel.ehrhold{at}ifremer.fr.
A 200-km2 area in the Bay of Concarneau on the South Brittany coast was surveyed acoustically using different sidescan sonars (a 100-kHz EdgeTech DF1000, and a 240-kHz Reson SeaBat 8101). The area corresponds to a sector of the REBENT network. It was selected for its physical and biological characteristics, reflecting the sedimentary heterogeneity and biological diversity of Brittany's coastal seafloors. The work presented here illustrates the methodology for mapping subtidal seabed habitats in the context of the network. Backscatter mosaics were produced covering 100% of the survey area. Extensive ground-truthing was carried out involving 93 Shipek grab samples and 25 drop-down video profiles. From interpretation of the acoustic facies, 40 biological soft-bottom stations were sampled using a Hamon grab to characterize macrobenthic communities (>2 mm). The results indicated considerable variation in backscatter responses in relation to high densities of macrobenthic species (Lithothamnion, Asterias, Haploops, Maldane, Ophiocomina), and a wide variety of substratum types present within a relatively small area. Dense biocenoses of maerl were accurately surveyed from 20-m to <5-m depth (Lower Astronomical Tide; LAT). Boundaries of Haploops communities are associated with dense small pockmarks in the centre of the bay. The relationships between sediment sometimes colonized by macrobenthic species and backscatter responses are discussed.
Keywords: benthic habitats, grab sampling, maerl, ortho-rectified aerial photography, pockmark, sediment, sidescan sonar, submarine video
Received 3 June 2005; accepted 28 June 2006.
 |
Introduction |
|---|
 Top Introduction The REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
The mapping of seabed habitats is increasingly recognized as an important tool for both marine resource management and scientific study. Many government organizations managing coastal resources have progressed in defining habitats and developing standards for seafloor and benthic habitat mapping (Robinson et al., 1996; Starr et al., 1997; Foster-Smith et al., 1999; Davies et al., 2001; Kostylev et al., 2001; Degraer et al., 2002). The monitoring of macrobenthic habitats fills a gap in coastal marine environment knowledge and evaluates how vulnerable marine habitats are to anthropogenic disturbance, such as trawling (Kenchington et al., 2001; Humborstad et al., 2004), commercial aggregate dredging (Brown et al., 2002; Limpenny and Meadows, 2002), or accidental pollution (Wahle et al., 1999). Ifremer launched and coordinated the development of a strategy for the REBENT network (REseau BENThique) in 2000 to monitor the aftermath of the "Erika" oil spill in December 1999, followed by that of MV "Prestige" in November 2002, in France (Guillaumont et al., 2002). The REBENT network will provide consistent baseline knowledge about coastal benthic habitats and constitute a monitoring tool to detect changes at various scales over time and space, particularly in terms of spatial extension or regression, and biodiversity (Guillaumont et al., 2002). Sector-based, seabed habitat mapping is currently being conducted throughout Brittany's coastal waters through a combination of geoacoustic marine systems (sidescan sonar, multibeam echosounders) and ground-truthing, using seabed video, grab sampling, and sediment analysis. In shallow water (<50 m deep), seabed sidescan sonar resolution and the 10–15-km2 surface area covered per day means that sidescan sonar remains the main remote acoustic system used to identify bedforms or biogenic accumulations (Blondel and Murton, 1997; Kenny et al., 2003). Acoustic imaging devices play a significant role in revealing much about benthic species' organization, behaviour, and interactions between themselves and the sedimentary habitat in which they live (Solan et al., 2003). The texture and the strength of the acoustic return of the sidescan record itself (sonogram) can predict or can be used to estimate the sedimentological areal extent and boundaries of dense biocenoses like maerl (Birkett et al., 1998), molluscs (Crepidula fornicata, Ehrhold et al., 1998; Crassostrea sp., Smith et al., 2001; Placopecten magellanicus, Kostylev et al., 2003), echinoderms (Dendraster excentricus, Fenstermacher et al., 2001), or seafloor vegetation (Zostera marina, Blondel and Murton, 1997; Posidonia oceanica, Siljestroem et al., 1996). Processed sidescan sonar, shallow multibeam echosounder, and grab samples supplemented by seabed video records are currently used to define marine habitat maps in the REBENT network. Here we deal only with the sidescan sonar results from the subtidal, sector-related approach in the Bay of Concarneau in southern Brittany (47°50'N 03°56'W, Figure 1). We present REBENT methodology and the potential for sidescan sonar to detect with accuracy distinct biocenoses, and discuss the difficulties in calibrating and fully understanding the spatial variability detected using the technique.
|
|
The REBENT network |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
In order to respond to increasing demand from the international community for knowledge and monitoring of coastal benthic biocenoses (Habitats Directive, 1992 and Water Framework European Directive, 2000, Natura 2000 network), the REBENT network is forming strong international partnerships to develop a framework for the European Network of Excellence in Marine Biodiversity (MARBEF), and to harmonize and optimize the Mapping of European Seabed Habitats (MESH; Connor and Coggan, 2004). Its aim is to define a reference state and to provide regular monitoring of the French marine coastal fauna and flora from the intertidal to water depths of about 30 m. Before transposing the network strategy onto other seafronts (eastern Channel, Bay of Biscay, Mediterranean) and adapting it to regional specificities, the Brittany coast was chosen as a pilot region to test protocols and standard methods for marine habitat mapping. REBENT's spatial organization comprises various interlocking levels embedded on three levels at different scales, with increasingly frequent monitoring and coverage to detect changes in specific habitats (Figure 2). These involve regional, broad-scale habitat mapping for a general area, fine-scale habitat mapping for sector habitat (6-year frequency), and biannual monitoring at stations using a Smith grab (Hily and Grall, 2003). Our approach to seafloor habitat mapping involved two successive surveys. First, each sector (Figure 2b) was investigated in spring to collect geophysical data. High resolution, validated acoustic surveys contribute towards better understanding of the benthic community distribution and the monitoring of changes. Acoustic mapping technology (sidescan sonar and multibeam echosounder) is used for morpho-sedimentary or biological habitat discrimination, or both approaches (Smith and Greenhawk, 1998; McRea et al., 1999; Cochrane and Lafferty, 2002; Ojeda et al., 2004), and can be effectively calibrated by ground-truthing (grab sample, submarine video, diving). Second, following the geophysical survey, we conducted stratified sampling of macrobenthic communities from morpho-sedimentary units or distinct biocenoses (Kenny et al., 2003). This approach made it possible to reduce significantly the biological sampling carried out in the autumn in the morpho-sedimentary units determined during the first survey. By comparing biological data with the morpho-sedimentary envelopes, marine habitat maps are obtained. The year 2003 was devoted to testing the methodology, especially with regard to the biological characterization of the seabed habitats explored. The REBENT subtidal approach will be complemented in the intertidal using a specific methodology (Bonnot et al., 2005) that combines littoral orthophotographs, field validation, and elevation data provided by lidar (light detection and ranging).
|
|
Material and methods |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
Survey locations and dates
The preliminary geophysical survey was conducted in March and October 2003 over an area of approximately 200 km2 in the Bay of Concarneau on the southern coast of Brittany (Figure 3). The bay stretches from the southern Breton coast between the Mousterlin and Trévignon headlands to the Glénan Islands. The seafloor slopes steadily southeastwards to a depth of 30 m, except on its eastern boundary, where the coastal topography features a terrace of shoals, the foot of which is bounded by the 20-m (LAT) isobath and orientated in the direction of the Kerforne fault system (Delanoë and Pinot, 1977). The bay is a Tertiary trough fault, 5–6 km wide and about 15 km long, bounded on the east by this major structural feature, and to the west by a parallel fault (Menier, 2003). The area was chosen because it exhibited a variety of seabed types, combined with complex nearshore bathymetry and specific biocenoses of maerl previously mapped. This was revealed from ground-truthing, diver observation, and grab-sample surveys (Grall and Hall-Spencer, 2003). The survey extended from depths of about 30 m to 10 m below the RV "Thalia" in March 2003, and was taken closer inshore in October 2003 using the small "Survex" craft (a mobile acoustic platform for shallow water).
|
Instrumentation and strategy
In the REBENT geophysical survey two independent acoustic systems were deployed (Table 1): a towed EdgeTech DF1000 sidescan sonar, and a hull-mounted Reson SeaBat 8101 multibeam sidescan sonar-based system. For the Bay of Concarneau, 400 km of profiles were collected using the EdgeTech sidescan between 4 and 13 March 2003 (Figure 1). Layback was recorded automatically every second. Position accuracy was ±5 m, including DGPS navigational error. Lane spacing between successive vessel tracks was 180 m, overlapping by 30% to ensure a complete mosaic (Figure 3). The EdgeTech DF1000 sidescan sonar is a dual-frequency system operating at 100 kHz, "standard resolution", and 500 kHz, "high resolution". The sonar signals are digitized in the towfish and transmitted via the tow cable to the topside acquisition system. Only the 100 kHz backscatter signal of the sidescan sonar data was interpreted, because this lower frequency provides a large spectrum of acoustic facies with respect to the variety of seabed grain size, bedform, and biological features. Numerical data from the EdgeTech system were displayed and recorded using a Triton-Elics International Isis Sonar data-acquisition system. All sidescan sonar data were recorded digitally on DVD ROM for post-survey processing, copying, and storage. Hard copies of the 100-kHz sonograms were printed on thermal film during acquisition using an EPC HSP-100 thermal plotter. The same sidescan acquisition settings were used on each site in the first survey: 110 m for swath width, vertical beam width tilted down 20° from the horizontal, constant dynamic contrast value. At a maximum vessel speed of 5 knots, this configuration yielded a theoretical along-track resolution of
15 cm. Port and starboard sidescan sonar data were displayed in real time on a high-resolution colour display (1280 x 1024 pixels). Telemetry from the navigation systems and the towfish, including full towfish attitude information, was displayed in separate windows.
|
The first survey was completed in the nearshore zone during autumn 2003 using the SeaBat 8101 deployed from the "Survex", to test the REBENT strategy. Only the shallow coastal waters (depths ranging from 10-m (LAT) to about 1 m) between Trévignon Point and Concarneau harbour were investigated, and 300 km of profiles were collected (Figure 1). The 240-kHz Reson SeaBat 8101 multibeam echosounder with a sidescan sonar option is hull-mounted on a small aluminium "Survex" vessel adapted for geophysical surveys (MesurisST). The "Survex" measures about 8.6 m long by 3 m wide, with a small draft of 0.8 m, including the sonar transducers. Real time, multibeam operator control ensured very high resolution and accurate bathymetry (z < 0.05 m) and digital seabed acoustic backscatter (x, y < 0.1 m). The bathymetry was displayed and recorded simultaneously with the sidescan sonar imagery using QINSy 7 software. The swath width was about 80 m maximum, but generally ranged from 40 m to 60 m. The Applanix POS-MV 320 inertial navigator system ensured real-time accuracy of the system (vessel attitude corrections), including position, combined with a Leica 530 RTK Differential GPS. The EdgeTech DF1000 mosaic was completed by "Survex" data for nearshore Concarneau to Trévignon Point. Usually, data were collected at a boat speed of about 5 knots, but sometimes this was less when entering nearshore rocky shoals. The average surface area fully covered by "Survex" is 3 km2 d–1.
During typical at-sea operations, sidescan sonar data were processed, mosaicked, and ultimately used as a basis for choosing sediment sample locations. This ground-truth component of cruise operations was critical in accurately interpreting these data. Raw data (XTF files) were first processed to correct the geometric distortions inherent in the data from the Caraibes® Ifremer submarine mapping system. Caraibes® software is a mapping toolbox which processes data from multibeam echosounders and towed sidescan sonars. Sonograms were corrected (towed-fish depth and position, slant range correction, antenna gain) before being mosaicked using bespoke Caraibes® modules (e.g. interpolation, filtering, or contrast-enhancement functions). Though both sidescan sonar frequencies (100 and 500 kHz) were used to characterize the bottom, the backscatter mosaic was constructed from 100-kHz data, because they provided more horizontal range. The data were post-processed using two different pixel resolutions, 1 and 0.3 m. The mosaic with a pixel size of 1 m provides a rapid overview of the sector studied, showing acoustically distinct regions representing areas of different substrata, relief or dominant organisms and all the facies to be calibrated onboard (Figure 3). Mosaics with 0.3-m pixels retain the details of sedimentary dynamics such as small and medium subaqueous dunes or small biological aggregates. Mosaics were georeferenced in a GeoTIFF format (Tagged Image File Format), displayed in a GIS (Geographical Information System), and subsequently combined with data sets which included grab-sample description, underwater video interpretation, and photographs and samples of sediment and biota, to select the best sites for ground-truthing during survey operations. To create a GIS with a large amount of information, the different data were encoded into separate layers, and finished maps could be extracted from the GIS as combinations of any of the layers. Data on existing marine resources and mapping, surveying, and ground-truthing of the nearshore coastal zone were collected, then synthesized into a GIS database and a web-based system for dissemination. The data were stored in an ESRI ArcGIS database which will be distributed via an ArcIMS Website (www.ifremer.fr/REBENT).
Ground-truthing equipment
During the first survey in spring 2003, 93 grab samples were taken on the floor of the bay using a Shipek grab (Figure 1). This grab is shaped like a half cylinder, and can collect bed material over a sample area of 0.042 m2. During the second survey in autumn 2003 (Figure 4), 40 biological soft-bottom stations were sampled, using a 0.125-m2 Hamon grab to characterize macrobenthic communities (>2 mm). Sample locations were guided by the results of the geophysical surveys carried out earlier in the year. Given the accuracy of the DGPS system and intrinsic operational errors in the positioning of the grab, the precision of the sample location is within 20 m. Onboard, a quick textural and biological description backed up by digital photography was made in a scientific computerized logbook CASINO+® (Ifremer real time, data-acquisition software). CASINO+® acquires, displays, edits, and stores metadata parameters for the survey and the ship (scientific and bridge data) in an EXCEL table. One subsample (about 500 g) of sediment material was taken, put into a plastic bag and frozen. About 200 g of this sediment subsample was analysed for particle-size distribution. The samples were first wet-sieved on a 50-µm stainless steel sieve. The >50-µm fraction was dry-sieved using a sieve shaker, on a range of stainless steel sieves placed at –4 phi intervals, down to 4 phi (63 µm). The classification by Larsonneur et al. (1982), adopted for English Channel sedimentological mapping, was used to define sediment-type classes, and these were correlated to the EUNIS habitat classification (European Nature Information System; Davies et al., 2004). In all, 25 underwater video transects were collected in March to ground-truth acoustic signatures and in October 2003 to characterize habitats using a frame and towed sledge, bottom video-acquisition system. A very sensitive black and white camera (SIMRAD OSPREY 0.1 lux) was used without light to reduce scattering effects from suspended matter. A time-stamp synchronized with the DGPS time was overlain on one of the camera records to allow video georeferencing. Videotapes were sampled and coded every 10 s. Video records were then displayed and interpreted directly in Arcview® GIS, using various aspects of the ADELIE® Ifremer software.
|
|
Sidescan sonar data interpretation |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
Sidescan sonar data represent backscatter received by the transducers from an insonified region of seafloor. Acoustic backscatter is thought to be a function of the angle of incidence of the acoustic wavefront to the seafloor, surface roughness, impedance contrast across the sediment–water interface, topography, and volume reverberation (Urick, 1983). In digital sidescan sonar imagery, high backscatter is generally represented by dark tones, low backscatter by light tones. Generally, areas of high backscatter are associated with relatively coarser-grained sediments, and areas of low backscatter with relatively finer-grained sediments. The backscatter patterns were compared visually in order to establish the initial habitat areas on the sidescan sonar mosaic. In order to interpret the mosaics generated (Figure 3), acoustic facies classed by their grey level intensity and image texture were given one of four types (Table 2): (I) very low (white tones); (II) low (light grey tones); (III) moderate (dark grey tones); and high (very dark grey tones).
|
The spatial distribution of acoustic responses (Figure 3) shows a dual gradient increasing from the central part of the bay towards the edges of the basin. The backscatter intensity is moderate to low along the axis of the bay, becoming dark to very dark on the shoreface terrace between Beg Meil and Mousterlin, and on the edges of the Moutons bed. Particle-size analysis alone cannot satisfactorily explain this variability between shades of grey and roughness. In detail, the wide spectrum of acoustic responses covers homogeneous, flat seafloors, characterized by the presence of acoustically distinct biocenoses, small to medium dunes, and highly uneven seafloors. However, to interpret the geology of the area accurately, correlations must be made between the surficial sediment characteristics acquired by sidescan sonar and ground-truth techniques. A map (Figure 5) depicting the distribution of sedimentary environments in the study area was created from the sidescan sonar mosaic and grain size data using the dominant sediment type.
|
Very low backscatter intensity (class I, Table 2, Figures 3 and 4)
The homogeneous white facies (class I1, Figure 4) is limited to some small depressions along the eastern shore of the bay and some larger beds at the foot of the littoral slope at depths of 20 m (LAT) between Jument and Trévignon Points. Grab samples brought up a soft mud with a thickness of several centimetres of fluid mud on its surface. The pure mud (83% particles smaller than 50 µm) is particularly homogenous and lacking in macrofauna. This sediment, whose presence was not mentioned in earlier studies (Glémarec, 1973; Pinot, 1974), is probably temporary deposits of suspended particles (Jouanneau et al., 1999) released from the nearest estuaries (i.e. the Vilaine and the Loire). The cloudy white facies (class I2, Figure 4) is distributed in exactly the opposite way between silted-sandy bottoms at the entrance of the La Forêt Bay around –10 m deep (LAT), and the rippled fine-to-medium sandy bottoms are found in folds around the Moutons Island bed. This facies is not very different from the previous one in grain size terms. Samples taken at the entrance to La Forêt Bay showed the same features of freshly deposited mud, overlapping reduced and highly odorous sandy mud. To the east of the Moutons area, video and samples showed that this was clean, compact, fine-to-medium sand, sometimes slightly muddy (class I2, Figure 4) on the eastern fringe of this facies, but soft, and well sorted with more shells on its western fringe (class I3, Figure 4). At 1.5 nautical miles northeast of the Moutons on a sand substratum of 0.84 km2, sonograms (mottled facies I3) show relatively circular dark spots ranging from a few metres to about 50 m in diameter (Figure 6a). Video profiles and photographs taken by divers during the second survey in October 2003 show a high density of brittle stars (Amphiura filiformis), and occasional aggregates of sea stars (Asterias rubens) in the area. On the basis of the videos recorded in October, these acoustic spots cannot be linked with certainty to sporadic and high concentrations of sea stars, but large quantities of sea stars (Marthasterias glacialis, Asterias rubens), reaching 1 t d–1 boat–1, were regularly caught in the 1980s in this sector (Guillou, 1985).
|
Low backscatter intensity (class II, Table 2, Figures 3 and 4)
This area is a thin strip perpendicular to the coast between Beg Meil and Mousterlin, which widens offshore towards the centre of the bay. The homogeneous facies (class II1, Figure 4) corresponds to a flat seabed of compact, very fine, homogeneous sand (50% of sediment weight) mixed with clumps of compact mud (24%), which resists penetration by the Shipek grab. Biological samples taken using the Hamon grab in October 2003 showed that this was a thin sandy film covering sandy mud. The mottled facies (class II2, Figure 4), covering two surface areas of 2.5 and 1.7 km2, respectively, features alternating elongated dark patches running NW–SE parallel to the maximum currents (0.5 knots) in the zone (HW + 3 or LW + 3). They are about 10–20 m wide, with anastomosis in places (Figure 6b). These highly backscattering strips seem to be associated with high concentrations (up to 25 individuals 0.1 m–2; Ménesguen, 1980) of a very wide-tubed polychaete worm (Maldane glebifex). The feature increases surface roughness and modifies the sediment's physical properties.
Moderate backscatter intensity (class III, Table 2, Figures 3 and 4)
The grain size response that corresponds to this level of backscatter is complex, because field observations describe sediments ranging from sandy mud to well sorted and washed, coarse, ochre-coloured sand. Its homogenous aspect (class III1) is shown by a uniformly flat seabed that stretches from the centre of the bay all the way to the rocky littoral at Beg Meil. This fairly well-trawled area is made up of more compact mud, reduced in deeper areas, with a little clean very fine sand on its surface. The sedimentary envelopes farther south with the same grey level signature (class III2) are made up of sticky and compact mud, colonized in places by high concentrations of Haploops sp. (up to 860 individuals 0.1 m–2; Ménesguen, 1980). Video and diver observations (Figure 7b, c) showed up a true carpet of Haploops, with the exception of the centre of the craters (Figure 7a, b) where almost no colonies were present. These small craters, which are a few metres to >20 m in diameter, correspond to a field of pockmarks (Hovland and Judd, 1988). Seismic profiles show acoustic turbidity facies related to a contemporaneous organic matter degradation in the infilling transgressive deposits of the Concarneau palaeovalley (Menier, 2003), as well as being observed in the bays of Penobscot (Gontz et al., 2002) and Belfast (Kelley et al., 1994). The last class (class III3) corresponds to a facies in direct contact with the Moutons maerl bed (Figure 4). Made up of medium-to-coarse, well-sorted and finely shell-based sand, it has a rippled texture and is relatively symmetrical with a wavelength <2 m and height <0.3 m. Video observations confirm the loose and mobile nature of this sediment, which is easily resuspended by currents channelled between the Mouton shoals and the Glénan Archipelago.
|
Strong backscatter intensity (class IV, Table 2, Figures 3 and 4)
This is mainly limited to the terrace between Concarneau and Trévignon to the east, and to the Moutons bed to the west. It characterizes the rocky shoals (class IV5) and maerl communities in the bay (class IV4). Maerl beds in coarse, clean gravel, and sand sediments develop a specific acoustic response (Figures 4c and 8a) that clearly distinguishes them from all other sedimentary facies. This vulnerable habitat is found at depths from 0 to 10 m, generally on the downstream side of shallow rocks, and has been mapped using acoustic surveys around Brittany for many years (Augris and Hamon, 1996). Until then, they had been delimited in the region using biological sampling means (Pinot, 1974; Grall, 2002). They were very accurately surveyed in the bay, even in very shallow depths, from the "Survex" geophysical measurement launch and coastal ortho-rectified photographs (Figure 5a). The numerous validations performed in March and October 2003 showed that this high backscatter signature corresponds to a thick maerl deposit, patterned by small and medium sand dunes of a few metres in wavelength (1–5 m) and some decimetres of height. Seismic reflectivity recordings suggest that the largest of these beds is from 7 to 10 m thick (Pinot, 1974). Their surface area varies considerably, from a few m2 to several km2, covering 9 km2 in all of the insonified areas. At Concarneau, they develop either on a muddy substratum, generally attached to the rocky shoals between Trévignon and Beg Meil, or on coarse seabeds east of the Moutons. Owing to wave action, the maerl is laterally dispersed onto the muddy host sediment, draping a thin layer over it (class IV3, Figure 8b). In a very limited area west of Trévignon, this rippled acoustic signature was somewhat altered, and corresponded to the significantly dense presence of brittle stars (Ophiocomina nigra) on a maerl bed.
|
|
Discussion |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
Backscatter signatures and sedimentological interpretation
The relationships between sediment and backscatter are not always clear. A multitude of sedimentological factors (e.g. grain size, volumetric heterogeneity, fine-scale roughness of surface sediment) and significant slope variation may play an important role in the acoustic response (Urick, 1983). The seabed sediments of the middle part of the bay show predominantly low and moderate backscatter, apparently related to the mud-to-sand transition. However, contrary to the usual correlations between backscatter and grain size, in this situation the moderate or high backscatter is associated with predominantly muddy sediments. The most noticeable inconsistencies to this presumption are the moderate backscatter values that correspond to sandy mud deposits associated with both high densities of Haploops sp. (class III3) and heterogeneous sediment (sandy mud), with mud bottom partially uncovered by maerl vestiges. Dense populations of macrobenthic species (maerl, Maldane glebifex, Haploops sp., Amphiura filiformis, Ophiocomina nigra, starfish, etc.) influence the geotechnical and geophysical properties of surficial sediment (Rowden et al., 1998). Yet roughness, which is known to have a strong influence on the acoustic response of the seafloor, is principally related to the meso- (metre) and microscale (centimetre) morphology of the surface of the seafloor (Jackson et al., 1996). Hence, significant roughness, such as dense pockmarks, will induce diffusive scattering of the acoustic wave and, thus, greater backscatter than predicted from a flat surface. In addition, bioturbation (Goff et al., 1999; Urgeles et al., 2002) and trawling (Humborstad et al., 2004) seem to be major factors influencing seabed roughness (as in class IV3, for example). Video footage and grab samples of the seabed indicate that the muddy surface is roughened by bioturbation, so that may be an important factor making muddy sediments acoustically more reflective. Subsurface heterogeneity is also a possible cause of backscatter variation. The acoustic energy of 100-kHz sidescan sonar systems can penetrate up to 50 cm in muddy sediment (less in sand), and can be scattered significantly by heterogeneities such as small void spaces (Mitchell, 1998). According to Goff et al. (2000), a small extra percentage of large grain (>4 mm) material, such as shell hash or maerl, can degrade the correlation between the backscatter and the grain size.
|
Conclusions |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
The REBENT monitoring programme is a French network for surveying benthic habitats. Its aim is to define a baseline reference state and propose a regular monitoring system for marine coastal fauna and flora. The network entered its operational phase around Brittany in 2003. The mapping of macrobenthic biocenoses is based on a combination of several acoustic sensors, especially sidescan sonar, in order to reduce significantly the biological sampling effort. The use of geophysical methods and validation techniques to determine marine habitats is now well established. The REBENT strategy involves the use of "fit-for-purpose" survey vessels, and tools to address particular operational limitations such as water depth. A sonar towfish, such as the EdgeTech DF1000 sidescan can map approximately 14 km2 d–1 of the seabed, while the shallow-water launch equipped with the Reson imager can map around 3 km2 d–1 under optimal conditions. The work has demonstrated that shallow and deeper water acoustic techniques are key components in an integrated approach to the preliminary characterization of subtidal habitats. The quality of the sidescan recordings and the heterogeneity of the Breton seafloors assume that considerable work is needed to calibrate the acoustic facies during the survey or very quickly thereafter. One problem encountered concerned high densities of mobile species on the seabed (Amphiura filiformis, Ophiocomina nigra, Marthasterias glacialis, Asterias rubens) whose signature can be wrongly interpreted as acoustic noise. Initial results obtained for the Bay of Concarneau highlight the progress made in our knowledge of morpho-sedimentary variability of the seafloor in this region, and above all in mapping maerl biocenoses with high Haploops density within a large pockmark field. Morpho-textural analysis of the reflectivity mosaics based on an adequate number of acoustic signature calibration observations supply the framework for characterizing habitats while limiting biological sampling. Once completed, the biological studies will allow the setting up of a definitive map of marine habitats in the sector. The combination of shipboard sensors, management of the various data batches in GIS, and the observation modes used (diving, drop-frame, or towed video) reduce errors in interpreting the sidescan sonar imagery, and optimize our interpretations of the seabed, even when they seem apparently homogeneous.
|
Acknowledgements |
|---|
Our thanks are due to the skipper and the crew of the Genavir survey vessels. The REBENT project was supported by the Diren Bretagne and the Total foundation.
|
References |
|---|
|
TopIntroductionThe REBENT networkMaterial and methodsSidescan sonar data...DiscussionConclusionsReferences |
|---|
-
Augris C. and Hamon D. (1996) Atlas thématique de l'environnement marin en baie de Saint-Brieuc (Côtes d'Armor)Edition IFREMER 72 pp.
Birkett D. A., Maggs C., Dring M. J. (1998) Maerl (volume V). An overview of dynamic and sensitivity characteristics for conservation management of marine SACs. Scottish Association for Marine Science (UK Marine SACs Project report.). 117 pp.
Blondel P. and Murton B.J. (1997) Handbook of Seafloor Sonar ImageryJohn Wiley, in association with Praxis Publishing Ltd 314 pp.
Bonnot C., Fournier J., Rollet C., Populus J., Guillaumont B., Loarer R. (2005) Bio-morpho-sedimentary cartography of the tidal zones in the French benthic network context. Photo-Interprétation 2:4125–27.
Brown C.J., Cooper K.M., Meadows W.J., Limpenny D.S., Rees H.L. (2002) Small-scale mapping of sea-bed assemblages in the eastern English channel using sidescan sonar and remote sampling techniques. Estuarine, Coastal and Shelf Science 54:263–278.[CrossRef]
Cochrane G.R. and Lafferty K.D. (2002) Use of acoustic classification of sidescan-sonar data for mapping benthic habitat in the Northern Channel Islands, California. Continental Shelf Research 22:683–690.[CrossRef][Web of Science]
Connor D. and Coggan R. (2004) Development of a framework for Mapping European Seabed Habitats (MESH). Abstracts of GEOHAB Marine Geological and Biological Habitat Mapping Fifth International SymposiumGalway, Ireland pp. 34–35.
Davies C.E., Moss D., Hill M. (2004) Eunis Habitat Classification Revised 2004Eunis Publication 307 pp.
Davies J., Baxter J., Bradley M., Connor D., Khan J., Muray E., Sanderson W., Turnbull C., Vincent M. (2001) Marine Monitoring Handbook (UK Marine SACs Project). 221 pp.
Degraer S., van Lancker V., Moerkerke G., Van Hoey G., Vincx M., Jacobs P., Henriet J. P. (2002) Intensive evaluation of the evolution of a protected benthic habitat: HABITAT. Final report of the Federal Office for Scientific, Technical and Cultural Affairs (OSTC) – Ministry of the Flemish Community, Environment and Infrastructure. Department. Waterways and Marine Affairs Administration, Coastal Waterways. 124 pp.
Delanoë Y. and Pinot J.P. (1977) Littoraux et vallées holocènes submergés en baie do Concarneau (Bretagne méridionale). Bulletin de l’Association française pour l’étude du quaternaire 3:27–38.
Ehrhold A., Blanchard M., Auffret J.P., Garlan T. (1998) The role of Crepidula proliferation in the modification of the sedimentary tidal environment in Mont-Saint-Michel Bay (The Channel, France). Comptes Rendus de l'Académie des Sciences, Series IIA, Earth and Planetary Science 327:583–588.[CrossRef]
Fenstermacher L., Crawford G.B., Borgeld J.C., Britt T., George D., Klein M., Driscoll N.W., Mayer L.A. (2001) Enhanced acoustic reflectivity due to high abundance of sand dollars, Dendraster excentricus. Marine Georesources and Geotechnology 19:135–145.
Foster-Smith R. L., Davies J., Sotheran I. (1999) Broad-scale remote survey and mapping of sublittoral habitats and biota. Report of sublittoral mapping methodology of the BMAP Project, Sea Map Research Group. 157 pp.
Glémarec M. (1973) The benthic communities of the European North Atlantic continental shelf. Oceanography and Marine Biology. An Annual Review 11:263–289.
Goff J.A., Olson H.C., Duncan C.S. (2000) Correlation of sidescan-backscatter intensity with grain-size distribution of shelf sediments. Geo-Marine Letters 20:43–49.
Goff J.A., Orange D.L., Mayer L.A., Hughes Clarke J.E. (1999) Detailed investigation of continental-shelf morphology using a high-resolution, swath sonar survey: the Eel margin, northern California. Marine Geology 154:255–269.[CrossRef][Web of Science]
Gontz A.M., Belknap D.F., Kelley J.T. (2002) Seafloor features and characteristics of the black-ledges area, Penobscot Bay, Maine, USA. Journal of Coastal Research 36:333–339.
Grall J. (2002) Biodiversité spécifique et fonctionnelle du maerl: réponses à la variabilité de l'environnement côtier. PhD thesis, University of Bretagne occidentale, France. 302 pp.
Grall J. and Hall-Spencer J.M. (2003) Problems facing maerl conservation in Brittany. Aquatic Conservation: Marine and Freshwater Ecosystems 13:S155–64.[CrossRef]
Guillaumont B., Archambault P., Bouvet G., Hamon D., Hily C., Derrien-Courtel S., Abiven T., Grall J., Le Hir M., Le Loc'h F., Raffin C., Castric-Fey A., Berline L., Dion P., Ballu S., Augris C., Bonnot-Courtois C., Fournier J., Ar Gall E., Singelin P., Pellarin Y., Lesage G. (2002) Réseau Benthique (REBENT). Développement d'un pilote breton. Elaboration de l'Avant-Projet Détaillé (APD) – Phase 1. Ifremer/Deal, Diren, Préfecture de la région Bretagne, Brest. 65 pp.
Guillou M. (1985) Dynamique des populations d'étoiles de mer en baie de Douarnenez (Finistère). Oceanis 11:253–261.
Habitats Directive. (1992) Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal of the European Communities OJ L 206(22/7/1992) 58 pp. + 5 Annexes.
Hily C. and Grall J. (2003) Suivi stationnel des bancs de Maerl. Fiche technique REBENT. www.ifremer.fr/REBENT/Document/. 4 pp. + 1 Annex.
Hovland M. and Judd A.G. (1988) Seabed Pockmarks and Seepages. Impact on Geology, Biology and the Marine Environment(Graham and Trotman, London) 293 pp.
Humborstad O.B., Nøttestad L., Løkkeborg S., Rapp H.T. (2004) RoxAnn bottom classification system, sidescan sonar and video-sledge: spatial resolution and their use in assessing trawling impacts. ICES Journal of Marine Science 61:53–63.[CrossRef][Web of Science]
Jackson D.R., Briggs K.B., Williams K.L., Richardson M.D. (1996) Test of models for high-frequency, seafloor backscatter. IEEE Journal of Oceanic Engineering 21:458–470.[CrossRef][Web of Science]
Jouanneau J.M., Weber O., Cremer M., Castaing P. (1999) Fine-grained sediment budget on the continental margin of the Bay of Biscay – critical interface on continental margins. Deep-Sea Research II, Topical Studies in Oceanography 46:2205–2220.[CrossRef]
Kelley J.T., Dickson S.M., Belknap D.F., Barnhardt W.A., Henderson M. (1994) Giant sea-bed pockmarks: evidence for gas escape from Belfast bay, Maine. Geology 22:59–62.
Kenchington E.L.R., Prena J., Gilkinson K.D., Gordon D.C., MacIsaac K., Bourbonnais C., Schwinghamer P.J., Rowell T.W., McKeown D.L., Vass W.P. (2001) Effects of experimental otter trawling on the macrofauna of a sandy bottom ecosystem on the Grand Banks of Newfoundland. Canadian Journal of Fisheries Aquatic Sciences 58:1043–1057.[CrossRef]
Kenny A.J., Cato L., Desprez M., Fader G., Schüttenhelm R.T.E., Side J. (2003) An overview of seabed-mapping technologies in the context of marine-habitat classification. ICES Journal of Marine Science 60:411–418.
Kostylev V.E., Courtney R.C., Robert G., Todd B.J. (2003) Stock evaluation of giant scallop (Placopecten magellanicus) using high-resolution acoustics for seabed mapping. Fisheries Research 60:2–3479–492.[CrossRef][Web of Science]
Kostylev V.E., Todd B.J., Fader G.B.J., Courtney R.C., Cameron G.D.M., Pickrill R.A. (2001) Benthic-habitat mapping on the Scotian Shelf based on multibeam bathymetry, surficial geology and sea-floor photographs. Marine Ecology Progress Series 219:121–137.[CrossRef][Web of Science]
Larsonneur C., Bouysse P., Auffret J.P. (1982) The superficial sediments of the English Channel and its western approaches. Sedimentology 29:851–864.[CrossRef][Web of Science]
Limpenny D.S. and Meadows W. (2002) Remote acoustic methods for examining seabed. In Boyd S.E. (Ed.). Guidelines for the Conduct of Benthic Studies at Aggregate Dredging Sites(UK Department for Transport, Local Government and the Regions, London and Cefas, Lowestoft) pp. 52–67 117 pp.
McRea J.E., Greene H.G., O'Connell V.M., Wakefield W.W. (1999) Mapping marine habitats with high-resolution sidescan sonar. Marine benthic habitats and their living resources: monitoring, management and applications to Pacific Island countries. Oceanologica Acta 22:679–686.[CrossRef][Web of Science]
Ménesguen A. (1980) La macrofaune benthique de la baie de Concarneau: peuplements, dynamique de populations, prédation excercée par les poissons. PhD thesis, University of Bretagne occidentale, France. 127 pp.
Menier D. (2003) Morphologie et remplissage des vallées fossiles sud-armoricaines: Apports de la stratigraphie sismique. PhD thesis, University oy Bretagne sud, France. 212 pp.
Mitchell N.C. (1998) Model for attenuation of backscatter due to sediment accumulations and its application to determine sediment thickness with GLORIA sidescan sonar. Journal of Geophysical Research 98:22477–22493.
Ojeda G.Y., Gayes P.T., van Dolah R.F., Schwab W.C. (2004) Spatially quantitative habitat mapping: example from the northern South Carolina inner continental shelf. Estuarine, Coastal and Shelf Science 59:399–416.[CrossRef]
Pinot J.P. (1974) Le précontinent breton, entre Penmarch, Belle-lle et l'escarpement continental, étude géomorphologique(Impram, Lannion) 248 pp.
Robinson C.L.K., Hay D.E., Booth J., Truscott J. (1996) Standard methods for sampling resources and habitats in coastal subtidal regions of British Columbia. 2. Review of sampling with preliminary recommendations. Canadian Technical Report of Fisheries and Aquatic Sciences XII:119.
Rowden A.A., Jago C.F., Jones S.E. (1998) Influence of benthic macrofauna on the geotechnical and geophysical properties of surficial sediment, North Sea. Continental Shelf Research 18:1347–1363.[CrossRef][Web of Science]
Siljestroem P.A., Rey J., Moreno A. (1996) Characterization of phanerogam communities (Posidonia oceanica and Cymodocea nodosa) using sidescan-sonar images. Photogrammetry and Remote Sensing 51:308–315.[CrossRef]
Smith G.F., Bruce D.G., Roach E.B. (2001) Remote acoustic habitat assessment techniques used to characterize the quality and extent of oyster bottom in the Chesapeake Bay. Marine Geodesy 24:171–189.[CrossRef]
Smith G.F. and Greenhawk K.N. (1998) Shellfish benthic-habitat assessment in the Chesapeake Bay: progress toward integrated technologies for mapping and analysis. Journal of Shellfish Research 17:1433–1437.[Web of Science]
Solan M., Germano J.D., Rhoads D.C., Smith C., Michaud E., Parry D., Wenzhöfer F., Kennedy R., Henriques C., Battle E., Carey D., Iocco L., Valente R., Watson J., Rosenberg R. (2003) Towards a greater understanding of pattern, scale and process in marine benthic systems: a picture is worth a thousand worms. Journal of Experimental Marine Biology and Ecology 285/286:313–338.
Starr R., Steger J., Greene H. G., Schwing F. (1997) Mapping Benthic Habitats and Ocean Currents in the Vicinity of Central California's Big Creek Ecological Reserve. NOAA Report, TM-NMFS-SWFSC, 245. 70 pp.
Urgeles R., Locat J., Schmitt T., Hughes Clarke J.E. (2002) The July 1996 flood deposit in the Saguenay Fjord, Quebec, Canada: implications for sources of spatial and temporal backscatter variations. Marine Geology 184:41–60.[CrossRef][Web of Science]
Urick R.J. (1983) Principles of Underwater Sound(McGraw-Hill Book Company, New York) 423 pp.
Wahle R.A., Cobb J.S., Clancy M. (1999) Habitat-based assessment of lobster abundance: a case study of an oil spill. American Fisheries Society Symposium 22:285–298.[Web of Science]
Water Framework European Directive. (2000) Directive 2000/60/EC of the European parliament and the council of 23 October 2000, establishing a framework for Community action in the field of water policy. Official Journal of the European Communities OJ L 327(22/12/2000). 72 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||