Grab sampling

Review of standards and protocols for seabed habitat mapping February 2007 Review of standards and protocols for seabed habitat mapping Date Authors 2005/03 R. COGGAN, M. CURTIS, S. VIZE C. JAMES S. PASSCHIER A. MITCHELL 2006/03 2006/07 2006/09 2007/02 2007/02 B. FOSTER-SMITH J. WHITE S. PIEL, J. POPULUS V. VAN LANCKER S.PIEL K. VANSTAEN S. DELEU V. JEGAT K. SHEEHAN, F. FITZPATRICK J. WHITE J. WHITE R. COGGAN J. POPULUS Organisation CEFAS, United Kingdom BGS, United Kingdom TNO, The Netherlands Queen’s University of Belfast, Agri-food and Biosciences Institute, United Kingdom. Envision Ltd, United Kingdom Marine Institue, Ireland IFREMER DYNECO/VIGIES, France University of Ghent, Belgium IFREMER DYNECO/VIGIES, France CEFAS, United Kingdom University of Ghent, Belgium Marine Institute, Ireland Marine Institute, Ireland Marine Institute, Ireland Marine Institute, Ireland CEFAS, United Kingdom IFREMER DYNECO/VIGIES, France Version 1.0 1.1 1.2 1.3 1.4 CONTENT This document consists of 210 pages including cover and contents. This document was published within the framework of: MESH Action 2.1, which is an INTERREG lllB-NW European Program. http://www.searchmesh.net/ Reference to this report should be cited to individual Chapter and their Authors and to the edited volume as: | † ‡ ‡ ‡ † Coggan, R ., Populus, J ., White, J ., Sheehan, K ., Fitzpatrick, F . and Piel, S . (eds.) (2007). Review of Standards and Protocols for Seabed Habitat Mapping. MESH. | CEFAS, UK Ifremer, France ‡ Marine Institute, Ireland † Review of standards and protocols for seabed habitat mapping CONTENTS MESH Programme .................................................................................................................................. 1 REMOTE SENSING TECHNIQUES 1 High Resolution Satellite Imagery................................................................................................ 2 Steven Piel and Jacques Populus (Ifremer) 1 – General Principles of Operation ..................................................................................................... 2 2 – Variety of Systems Available.......................................................................................................... 2 3 – Review of Existing Standards and Protocols ................................................................................. 5 4 – Current Usage ................................................................................................................................ 8 2 Airborne Digital Imagery ............................................................................................................. 11 Steven Piel and Jacques Populus (Ifremer) 1 – General Principles of Operation ................................................................................................... 11 2 – Variety of Systems Available........................................................................................................ 12 3 – Review of Existing Standards and Protocols ............................................................................... 13 4 – Provenance and Current Usage .................................................................................................. 15 3 Aerial Photography...................................................................................................................... 17 Steven Piel and Jacques Populus (Ifremer) 1 – General Principles of Operation and Data Processing ................................................................ 17 2 – Variety of Systems Available........................................................................................................ 17 3 – Review of Existing Standards and Protocols ............................................................................... 18 4 – Provenance and Current Usage .................................................................................................. 19 4 Lidar .............................................................................................................................................. 21 Steven Piel and Jacques Populus (Ifremer) 1 – General Principles of Operation and Data Processing ................................................................ 21 2 – Variety of Systems Available........................................................................................................ 24 3 – Review of Existing Standards and Protocols ............................................................................... 28 5 Planning Considerations for Remote Sensing of Benthic Habitats ....................................... 32 Steven Piel and Jacques Populus (Ifremer) 1 – Data Requirements versus Spatial Scales................................................................................... 32 2 – Temporal Scales .......................................................................................................................... 35 3 – Ground Truthing Techniques to Validate Remote Sensing Data................................................. 36 4 – Positioning systems ..................................................................................................................... 38 5 – Planning Considerations .............................................................................................................. 38 6 – Synergy between Lidar and Remote Sensing Imagery ............................................................... 39 7 – Cost Considerations..................................................................................................................... 41 6 Sea Bottom Topography with Navigation Radar ...................................................................... 44 Jan Kleijweg (TNO-FEL) 1 – General Principles of Operation ................................................................................................... 44 ACOUSTIC SYSTEMS TECHNIQUES 7 Sidescan Sonar ............................................................................................................................ 45 Ceri James (BGS) 1 – Principles of Operation and Data Processing .............................................................................. 45 2 – Sidescan Systems........................................................................................................................ 47 3 – Review of Existing Standards and Protocols ............................................................................... 48 8 Multibeam Echo Sounders.......................................................................................................... 53 Jonathan White and Veronique Jegat (Marine Institute), Vera Van Lancker, Samual Deleu (Ghent University) and Koen Vanstaen (CEFAS) 1 – General Principles of Operation and Data Processing ................................................................ 53 2 – Technical Details .......................................................................................................................... 54 3 – Data Acquisition ........................................................................................................................... 59 4 – Backscatter Data.......................................................................................................................... 63 5 – Evaluation of Multibeam Echosounders for Habitat Mapping Purposes...................................... 64 Review of standards and protocols for seabed habitat mapping 6 – Review of Existing Standards and Protocols ............................................................................... 65 7 – Provenance and Current Usage .................................................................................................. 69 9 Interferometric sonar systems ................................................................................................... 73 Jonathan White and Veronique Jegat (Marine Institute) 1 – General Principles of Operation and Data Processing ................................................................ 73 2 – Varieties of System Available....................................................................................................... 77 3 – Review of Existing Standards and Protocols ............................................................................... 78 4 – Provenance and Current Usage .................................................................................................. 80 10 Acoustic Ground Discrimination Interpreted With Ground Truthing ..................................... 83 Bob Foster-Smith (Envision Ltd.) 1 – General Principles of Operation and Data Processing ................................................................ 83 2 – General Principles of the Technique............................................................................................ 83 3 – Using AGDS ................................................................................................................................. 84 4 – Review of Existing Standards and Protocols ............................................................................... 86 11 Seabed Imaging Using Existing 3D Marine Exploration Seismic Data Sets.......................... 94 Joe Bulat (BGS) 1 – General Principles of Operation and Data Processing ................................................................ 94 2 – Varieties of System Available....................................................................................................... 95 3 – Review of Existing Standards and Protocols ............................................................................... 95 4 – Provenance and Current Usage .................................................................................................. 97 12 Sub-Bottom Acoustic Profiling .................................................................................................. 99 C. Mesdag (TNO-NITG) 1 – General Principles of Operation and Data Processing ................................................................ 99 2 – Varieties of System Available..................................................................................................... 101 3 – Review of Existing Standards and Protocols ............................................................................. 102 4 – Provenance and Current Usage ................................................................................................ 104 IN SITU SAMPLING TECHNIQUES 13 Diver Surveys ............................................................................................................................. 106 Annika Mitchell (Queen’s University of Belfast) and Neil Golding (JNCC) 1 – General Principles of Operation and Data Processing .............................................................. 106 2 – Data Acquisition ......................................................................................................................... 108 3 – Data Processing ......................................................................................................................... 110 4 – Data Interpretation ..................................................................................................................... 111 5 – Provenance and Current Usage ................................................................................................ 112 6 – Summary of Future Guideline Recommendations ..................................................................... 113 14 Particle Size Analysis (granulometry) of Sediment Samples................................................ 116 S. Passchier (TNO-NITG) 1 – General Principles of Operation and Data Processing .............................................................. 116 2 – Varieties of System Available..................................................................................................... 117 3 – Review of Existing Standards and Protocols ............................................................................. 118 4 – Provenance and Current Usage ................................................................................................ 122 15 Trawls and Dredges................................................................................................................... 127 Samantha Vize and Roger Coggan (CEFAS) 1 – General Principles of Operation and Sample Processing.......................................................... 127 2 – Georeferencing and General Information .................................................................................. 127 3 – Sample Processing .................................................................................................................... 128 4 – Variety of Systems Available...................................................................................................... 129 5 – Review of existing standards and protocols............................................................................... 132 6 – Data Interpretation ..................................................................................................................... 134 7 – Provenance and Current usage ................................................................................................. 135 8 – Recommendations ..................................................................................................................... 136 16 Geotechnical Measurements .................................................................................................... 139 Fiona Fitzpatrick (Marine Institute) and Dave Long (BGS) 139 1 – Introduction and bulk sediment properties ................................................................................. 139 2 – Description of the sediment ....................................................................................................... 141 Review of standards and protocols for seabed habitat mapping 3 – Measurement techniques........................................................................................................... 146 17 Grab Sampling ........................................................................................................................... 152 Andrew S. Y. Mackie (NMW), Roger Coggan (Cefas) and Sytze van-Heteren (TNO) 1 – General Principles of Operation and Sample Processing.......................................................... 152 2 – Requirements for a Quantitative Benthic Sampler..................................................................... 152 3 – Overview of Current Grab Operation ......................................................................................... 153 4 – Variety of Systems Available...................................................................................................... 156 5 – Review of Existing Standards and Protocols ............................................................................. 159 6 – Provenance and Current usage ................................................................................................. 163 VIDEO AND IMAGERING TECHNIQUES 18 Sediment Profile Imagery.......................................................................................................... 171 Matt Curtis and Roger Coggan (CEFAS) 1 – General Principles of Operation ................................................................................................. 171 2 – Varieties of Systems Available................................................................................................... 172 3 – Review of Existing Standards and Protocols ............................................................................. 173 19 Remote Video techniques......................................................................................................... 179 Annika Mitchell (Queen’s University of Belfast) and Roger Coggan (CEFAS) 1 – Introduction ................................................................................................................................ 179 2 – General Principles of Operation and Data Processing .............................................................. 179 3 – Varieties of System Available..................................................................................................... 182 4 – Review of Existing Standards and Protocols ............................................................................. 188 5 – Spatial Positioning and Georeferencing of Information ............................................................. 191 6 – Video Analysis and Interpretation .............................................................................................. 193 7 – Data Storage .............................................................................................................................. 195 8 – Data Interpretation ..................................................................................................................... 196 9 – Overall Evaluation of Existing Standards and Protocols............................................................ 198 10 – Common Recommendations.................................................................................................... 198 Review of standards and protocols for seabed habitat mapping 17 Grab Sampling Andrew S. Y. Mackie (NMW), Roger Coggan (Cefas) and Sytze van-Heteren (TNO) The benthic macroinfaunal invertebrates are considered a useful group to study in species assemblage mapping and environmental monitoring studies. This is because the majority of species are essentially sedentary and their natural distributions usually show good relationships with their sedimentary habitat and depth. Their responses to environmental change can easily be measured. They are an integral part of marine food webs and can be an important source of food for certain commercially exploited fish and invertebrates. More practically, the taxonomic literature on the worms, crustaceans, molluscs and echinoderms that are the main components of the macrofauna is generally good. Finally, the ‘soft-bottom’ benthos can be readily sampled by grabs, dredges and trawls. This review considers the use of grabs in quantitative assessments of seabed habitats, and concentrates on four types, namely the van Veen, Smith-McIntyre, Day and Hamon grabs. (Qualitative sampling by dredges and trawls is dealt with in a sister review, Vize and Coggan 2006). 1 – General Principles of Operation and Sample Processing Grabs are used to obtain quantitative samples of infaunal invertebrates and the substrates in which they live. If sufficient samples are taken, statistical variability among species, abundance and biomass can be investigated, providing a powerful means to compare samples from different places and different times. Grabs should be capable of repeatedly taking samples of a constant standard area – 2 2 2 nowadays commonly 0.1 m , though smaller (0.05 m ) and larger (0.2 m ) devices can be used. They should adequately sample the infauna contained below the area covered. An ideal sampler would routinely collected undisturbed sediment to a depth of 20 cm or more to capture all the infauna, including the larger, deep-dwelling, animals. The nearest device available is a large spade corer, such as the Reineck box corer (Reineck 1958, 1963; Farris and Crezee 1976) that can sample sediment to about 45 cm. However, such corers are large, heavy and require relatively large vessels to deploy them. Smaller sampling devices are therefore something of a compromise. Most of the smaller and most abundant infaunal species are present in the upper layers (5-10 cm) of the sediment. For this reason, a minimum volume of 5 litres sediment collected (equivalent to a sampling depth of 5 cm for a Van Veen grab) is regularly cited in sampling procedures (Kingston 1988, Riddle 1989b, Rumohr 1999). 2 – Requirements for a Quantitative Benthic Sampler Mackie (1981) summarised the basic requirements for a quantitative benthic sampler. The sampler should have: • • • • • • • • • • A minimum of working parts and be corrosion resistant. Be sturdy enough to withstand repeated deck handling and bottom impact. A bulk or weight that allows safe operation. The correct orientation on the seabed for sample collection. A trigger release mechanism to ensure actuation on the seabed at the proper time. A constant sample area. Adequate penetration of the sediment to capture the animals present. A low resistance to water on descent to minimise pressure-wave effects on surface-layer animals. Easy retrieval with no loss of sample. Easy removal of sample and quick redeployment capability. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 152 3 – Overview of Current Grab Operation The Smith-McIntyre, Day and Hamon grabs are all set in pyramidal frames (Figure 17–1). These give the grabs increased stability on the seabed, ensure correct orientation of the grab buckets, permit easy addition of supplementary weights and aid safe manual handling on deck. The ‘standard’ Van Veen grab lacks a frame and is more prone to incorrect orientation or toppling. A new modified grab (Mackie in prep, Mackie and Darbyshire 2001) has L-shaped bars attached to each bucket and its arm, effectively creating a frame. This L-frame Van Veen (Figure 17–2) is stable on the seabed and safer to manhandle and empty on retrieval. Figure 17–1. Hamon grab (University of Wales, Bangor). There are great similarities in the deployment of all the -1 grabs currently in use. Each is lowered (0.5-2.0 m.s ) vertically by wire via a winch and A-frame, jib or crane over the stern or side of the vessel. Although the grab is often lowered at constant speed, it may be allowed to free-fall for the last 5 m or so, to aid initial penetration (van Veen grab), or conversely, lowered more slowly to reduce the ‘bowwave’ effect (Smith-McIntyre, Day and Hamon grabs). A release mechanism is activated when the grab reaches the seabed and a sample is taken as the wire is hauled in. It is essential that the wire remains as vertical as possible on deployment and retrieval, otherwise the grab may by pulled to one side (‘grab-drift’) and only a partial (or no) sample may be collected (Kingston 1988, Riddle 1984). Figure 17–2. L-frame Van Veen grab, set ready for deployment (NMW). The ‘bow-wave’ effect displaces the often light or flocculent surface layer of many sediments, which can reduce the catch efficiency for small surface invertebrates – particularly microcrustaceans or surface-dwelling polychaetes (Smith and McIntyre 1954, Word 1976, Andersin and Sandler 1981). The incorporation of mesh screens in the top surface of the grab buckets (Fig. 3) can reduce this effect, allowing water to flow through the buckets during descent. The amount of mesh on the tops of grabs varies from 0-83% (e.g., Wigley 1967; Andersin and Sandler 1981). Rumohr (1990, 1999) recommended a minimum of 60%. Often, the mesh is itself covered by moveable flaps that act as one-way valves, allowing a through-flow of water on descent, but not on retrieval (Figure 17–3). Many grab designs also include sealable doors in to top of the grab buckets to allow inspection or sub-sampling of the undisturbed sample before the grab is emptied. Orton (1925) recognised that on final closure, the grab buckets can squeeze out lateral jets of water. This can be reduced by the use of mesh screens, or prevented by side guards (a feature of the Ponar grab; Powers and Robertson 1967), but these have rarely been used in marine surveys (Birkett 1958). All the grabs under consideration have the facility to add or remove weights to aid or limit penetrations Figure 17–3. Mesh screens with valveaccording to the type of sediment being sampled. On flaps, and inspection doors on the upper soft muddy sediments an excessively heavy grab will surface of a Van Veen grab. sink too far in and may not collect a representative sample. Conversely, on hard packed fine sand or coarse gravelly sediments, extra weight is required to keep the apparatus firmly in contact with the seabed. This weight acts against the initial gentle upward pull of the warp that causes the grab to close. Insufficient weight can lead to ‘warp-heave’, a slight raising of the grab above the seabed while it closes, reducing digging performance and sample size. Care is needed to add supplementary weights in a balanced way to minimise the risk of toppling the device. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 153 All designs are subject to certain factors that affect their performance. Riddle (1984) carried out extensive experiments into all aspects of grab efficiency, in laboratory and field situations. Kingston (1988) recognised five disadvantages associated with grabs activated or closed by the wire (warp). In addition to the warp heave, drift, initial penetration and the bow-wave effects mentioned above, the vertical movement of the ship in a swell can cause the grab to lose contact with the seabed (‘grabbounce’). All grabs have some form of trigger-release mechanism to hold the jaw(s) open as the grab descends. The trigger is normally activated on contact with the seabed. Where the mechanism relies on tension to be kept in the warp (Van Veen, and Hamon grabs), over-rapid deployment can cause the warp to become slack, allowing the trigger to fire during descent. Where the mechanism relies on mechanical friction and is physically triggered by contact with the seabed (Day and Smith-McIntyre grabs) ‘grabbounce’ may cause the trigger to fire early, or it may not fire at all if the sediment is soft or there is insufficient weight on the grab to overcome the friction. Weather conditions have a marked influence on triggering success, with failure rates increasing dramatically when sea conditions are greater than Beaufort force 5-6 (Riddle 1984, Kingston 1988, Kingston and Riddle 1989). Obviously, any large stones or shells that obstruct the closure of the jaws will cause the grab to fail. Even small obstructions can prevent full closure and lead to the sample being fully or partly washed out of the bucket as the grab is retrieved through the water column. In general, the ‘lighter’ Van Veen, Smith-McIntyre and Day grabs can be used from smaller vessels. A minimum vessel length of 10-12 m is usually sufficient to ensure enough deck space for operations and sample processing. The Hamon grab usually requires a larger vessel, on account of its greater weight (300-600 kg) and size. The size and number of samples taken depends largely upon the aims of the study (Green 1979; Riddle 1989a). Spatial and temporal replication must be carefully considered (Green 1979; Van der Meer 1997; Armonies 2000; Underwood and Chapman 2005). For a general grab sampling survey 2 designed to map the invertebrate assemblages of an area of seabed, a single 0.1m sample per station may be adequate (Cuff and Coleman 1979; but see also Green 1980, Cuff 1980). In other broad-scale studies (e.g., Mackie et al., 1995), at least two replicates have been taken to ‘even’ out the influence of any anomalous samples. In order to investigate animal-sediment relationships, samples of the sediment must be taken for particle size analysis (i.e. PSA / granulometry – see review by S. Passchier, this volume). Some studies collect sediment samples (approx 250 – 500 ml) from each grab used for faunal analysis. While this ensures that the sediment and faunal analyses relate to the same sample, this procedure can attract criticism from those who consider that it renders the sample semi-quantitative. Other studies take a specific replicate at each sampling station, for the sole purpose of granulometric analysis. However, this can attract the criticism that at heterogeneous stations, the sample used for PSA may differ significantly in its granulometry from the sample(s) taken for faunal analysis, leading to erroneous conclusions regarding the nature of animal-sediment relationships. 3.1 – Georeferencing and General Information The increasing accuracy and precision of positioning technology (dGPS), coupled with the fine detail now obtainable from remote sensing techniques such as multibeam sonar, has led to more rigorous recording of individual grab positions (see Rees et al., 1994). The grab is typically deployed from a location on the vessel that is some distance from the dGPS antenna, and this is often accounted for on larger vessels by recording/measuring the positional ‘offset’. Sometimes this fixed offset can be applied within the position recording software to automatically correct for the positional error. Occasionally, an acoustic tracking device is attached to the grab to record the exact sampling position, but this is not yet common practice The cruise plan and sampling logging system are usually agreed between the skipper and the scientist-in-charge before sampling begins. A test deployment of the grab is often made to ensure the gear and all recording systems are working properly. Metadata is recorded for each of the grabs to be used (e.g., type, design, modifications, weights etc) and for each sample taken. Sea-state and weather conditions are also commonly recorded. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 154 For sample acquisitions, common meta-data collected includes: date, time of deployment and retrieval of grab (alternatively time when the grabs hits the seabed), latitude, longitude, depth of water, depth of sample (e.g. in Day grab), sample volume (e.g. in for Hamon grab), visual sample status (valid/invalid sample) and visual sample assessment (e.g. the type, colour and smell of the sediment, the presence of shells, obvious faunal species and anthropogenic debris/litter). For sample processing, the metadata commonly records the purpose of the sample (i.e. for faunal or sediment analysis) and the size of the mesh through which the sample is sieved (1mm, 0.5mm). Each processed sample is labelled with sufficient information to record its unique identity. For example, MH0307/23B could mean “Milford Haven, July 2003 survey, Station 23, sample B”. Sample label writing can be minimised by pre-printing the standard survey information on waterproof labels, leaving only the station number and replicate to be added. When samples are separated into different sieved fractions (2 mm, 1mm, 0.5 mm), this is also recorded on the sample label (e.g. “2 mm, 4/5” could mean “2 mm sieve fraction, container 4 of a total of 5). Any fauna removed from the sample for individual fixation/preservation (e.g. scale-worms which are easily damaged) are similarly labelled. 3.2 – Sample Processing Sample processing is usually a 2-phase process. On the vessel, samples are sieved to remove the finer sediment, thus reducing the volume of material that has to be preserved and taken ashore. In the shore laboratory, the fauna are separated from the remaining sediment, identified, enumerated and weighed. For practical purposes the fauna of marine sediments are subdivided into three broad size-classes: meiofauna, macrofauna and megafauna. The separation of these groups is not well-defined and the size-distributions of the categories overlap. The meiofauna are generally considered to include animals passing through a 0.5 mm mesh sieve. Depending upon the meiofaunal taxon under investigation sieve meshes down to 30-40 µm may be used (McIntyre, 1969, McIntyre and Warwick, 1984). The macrofauna are usually considered to include those taxa retained on a 0.5 mm sieve, though in some situations larger or smaller meshes may be deemed necessary (e.g. see Bachelet, 1990). The megafaunal are very large animals that can be picked by hand. For studies of the shelf benthos, the standard sieve used has a minimum mesh size of 0.5 or 1.0 mm diameter. The latter is used largely for practical cost- and time-related reasons (Kingston and Riddle, 1989) as benthic macrofaunal processing can be very labour-intensive activity. Sieves with larger pore sizes (e.g. 2 mm, 5 mm) may be used above the standard sieve in order to separate mixed grade sediments into different size fractions, for more manageable processing on ship and in the laboratory. As grabs are retrieved on board, they are landed onto some kind of supporting frame where the collecting bucket or buckets can safely be opened (Figure 17–4). On most samplers, the success of the grab can be assessed by viewing the collected material through the inspection doors in the top surface of the grab (Figure 17–3). Samples are rejected if they do not meet certain pre-determined criteria, for example if the grab has leaked, if the surface of the sample is unduly disturbed, or if the volume is less than an acceptable minimum (commonly set a 5 litres, see Rumohr, 1999, Southern California Bight Field Methods Committee, 2002: Figure 3). Once on deck, the sample is usually washed over sieves to remove the finer material. A variety of sieve rigs are available, from large-scale washing baths (Eleftheriou and Moore, 2005; Figure 5.21) or tables (Rumohr, 1999: Figure 3), large trays (Figure 17–4), small hoppers (Eleftheriou and Moore, 2005: Figure. 5.19), sample tub–cradle–chute systems (Proctor et al., 2003: slide 42) to automated sieving devices like the Wilson Auto-Siever (http://www.tresanton.co.uk/wilson.shtml). A large metal ‘elutriator’ fitted with many spray nozzles (Figure 17–5) may be used as an alternative to the standard sieving methods, but these devices are not common. Whichever apparatus is used, copious amounts of sea-water and an abundance of large diameter sieves (e.g. 45 cm) are necessary for efficient elutriation. The crucial point of the sieving exercise is to extract the delicate invertebrate animals from the sediment in the best possible condition, as intact, well-preserved, animals are much easier to identify in the laboratory. Sample washing must therefore be a gentle rather than aggressive process. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 155 Figure 17–4. Washing and sieving grab samples. Figure 17–5. Elutriation machine (University of Wales Bangor/Ivor Rees). Once sieved, the different sample fractions are fixed using 8-12% formaldehyde (equivalent to 20-30% formalin) in seawater. This is two to three times the commonly recommended strength of formalin, but is necessary for adequate fixing of large volumes of sediment, which may also contain cryptic or tubedwelling species. The addition of a stain (usually Rose Bengal) to dye the fauna greatly aids the final processing. If the samples are to remain in formalin for some time, a buffer is usually added (hexamethylene tetramine or borax) to protect the shells of small bivalves while the samples await laboratory sorting (Rumohr, 1999). In the laboratory, samples are thoroughly washed in freshwater to remove the formalin, and the animals picked from the remaining sediment and preserved in 80% alcohol (with 2% propylene glycol added to reduce dehydration of the tissue). 4 – Variety of Systems Available A number of useful reviews of grab samplers have been published (Thorson, 1957; Holme, 1964, 1971; Hopkins 1964; Eleftheriou and Holme 1984; Kingston 1988, 2001, Rumohr, 1999; Eleftheriou and Moore, 2005). Good bibliographies are provided by Eleftheriou and Holme (1984), Elliott et al. (1993) and Eleftheriou and Moore, (2005). The following descriptions outline the construction and use of the four types of grab considered in this review. 4.1 – Van Veen Grab (van Veen 1933, 1936) See Figure 17–2 and Figure 17–3 (above), Eleftheriou and Moore 2005: Figure 5.11; Rumohr, 1999: Figure 1. The Van Veen grab comprises two quarter-circle buckets joined at a central pivot (Figure 17–3), each bucket having an overlapping arm. The arms may be long (Fig. 2) or short and are used to close the grab in a scissor-like movement. The arms are moved by chain or wire attached directly to their ends, or by a continuous warp threaded through pulleys on then end of each arm (as in Figure 17–2). The grab is used worldwide and particularly in Europe and North America. Typically, the Van Veen grab lacks a frame, though The National Museum of Wales have successfully operated several L-framed versions since 1997 (Mackie, in prep.; Figure. 2). The frame is attached to each arm, allowing easier handling and helping to prevent the grab from toppling onto its side when it lands on the seabed. This L-frame grab has about 30 % mesh area on the upper side of the buckets and also benefits from a bucket shaped for more efficient digging (Mackie, 1981: Riddle, 1984, 1989b). In the USA and Canada a variant (referred to variously as the Young Van Veen, Young grab, Ted Young grab) has relatively short arms and is set in the circular base of a ‘pyramidal’ frame. There are also variants with two grabs, side by side, in the same frame. The term ‘modified Van Veen’ used in various methods manuals and protocols (see below) does not refer to a single grab type. These publications however, usually contain a drawing, photograph or reference that can permit identification of the design Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 156 2 The Van Veen grab is relatively light, weighing 45-50 kg for a 0.1 m version, though lighter and 2 smaller (e.g. 0.05 m ) models are sometimes used on very soft substrata. The grabs can be weighted by adding lead to the arms, usually immediately above the buckets, though weights in excess of 100 kg tend to make the grabs top-heavy and more prone to toppling. Mackie (in prep.) found that when the grab was used on compacted mud, a small amount of weight was required on each arm to ensure proper firing of the trigger-release mechanism. The long-arm, continuous-warp version has the best digging efficiency compared to chain-rigged Van Veens, Day grabs and Smith-McIntyre grabs (Riddle, 1984, 1989b) tested on fine-medium sand. A 59 kg grab collected about 11 litres in tank tests. This is slightly more than the volume collected by heavier L-frame grabs in the field. Grabs of about 90-95 kg work reasonably well on harder bottoms and very well on most sands and all muddy sediments. The bucket shape of most Van Veen grabs means that the grab cannot fit in the hole it digs. On harder substrates, this causes the grab to lift slightly on closing and decreases the volume of sediment collected. The modified shape of the buckets in the L-frame Van Veen, and the novel Kingston Hydrostatic grab (Kingston 1988), removes this problem. The former has an initial penetration capability of 6.5cm, a maximum bite depth of 17 cm and can collect up to ~13 litres of material. 4.2 – Smith-McIntyre grab (Smith and McIntyre 1954) See Eleftheriou and Holme, 1984: Figure 6.15; Eleftheriou and Moore, 2005; Figure 5.12. The Smith-McIntyre grab (Figure 17–6) comprises two quarter-circle buckets mounted within a pyramidal frame (Fig. 6), giving greater stability to the grab compared to the Van Veen design. The buckets are connected to a central spring mechanism, which is compressed prior to deployment and released when two pressure plates hit the seabed, causing the buckets to be thrust into the sediment. On retrieval, warps attached to lever arms on each bucket cause them to close together, capturing a sample. The grab is popular worldwide but, due to the dangers presented with cocking the spring mechanism, the simpler Day grab is often preferred. Figure 17–6. Smith-McIntyre grab, ready for deployment (left) and on retrieval (right) © 2003 Kahl Scientific Instrument Corp., All Rights Reserved). 2 The Smith-McIntyre grab weighs 65-70 kg for a 0.1 m version. Typically, additional weights up to ~90 kg are added during normal operation The triggering of the spring mechanism may induce an upward movement of the frame on harder ground, reducing sampling efficiency, but this can be countered by 2 using extra weight. Bite profiles are semicircular and the 0.1 m version retains about 6 litres of sediment, (Riddle, 1984, 1989b). Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 157 4.3 – Day grab (Day 1978) See Eleftheriou and Holme 1984: Figure 6.16; Brown et al., 2002: Figures 7 and 8; Eleftheriou and Moore 2005; Figure 5.13. The Day grab (Figure 17–7) comprises two quarter-circle buckets mounted in a pyramidal frame having an 80-90 cm 2 square base (0.1m version). Each bucket has a short stubarm (centre of Figure 7) from which warps lead through the top of the fame, via pulleys on the base of the frame. The stub-arms also have a U-shaped lug on their inner surface, in which a trigger bar sits to hold the buckets open. At each end of the trigger bar, a downward extension ends in a horizontal plate that hangs a few centimetres below the base of the frame and cause the trigger bar to be lifted out of the Ushaped lugs when the grab is lowered to the seabed. Hauling on the warp then causes the buckets to close and collect a sample. Figure 17–7. Day grab prepared for Eagle et al. (1979) modified the flaps on the upper surface of deployment. Photo: Cefas. the buckets, changing the position of their hinges from the central axis to the outer upper edges of the grab. When the grab is retrieved, these flaps can be opened, allowing access to measure the depth of material retained and to take sub-samples, if required. Sturdy metal flaps are common, as simple rubber flaps can result in sample loss if the grab sinks too deeply into muddy sediments. 2 A typical 0.1 m Day grab weighs about 60-70 kg. As with other grabs, weights can be added to the frame as required. Commonly, in excess of 100 kg is used in normal operation, though Proudfoot et al. (2003) give 200 kg as the maximum weight. Insufficient weight can lead to the frame being pushed upwards as the buckets are drawn into the sediment, reducing the effective bite depth. Riddle (1984, 1989b) found bite profiles to be semicircular and about 7.4 l was collected on the test sediment. Brown et al. (2002) reported a maximum bite depth of 14 cm. The sturdy design, simple mechanism and ability to access the undisturbed surface of the sample make the Day grab a popular device for sampling marine benthos. It does not however, work well on hard, coarse, substrata. 4.4 – Hamon grab (Oele 1978) See Eleftheriou and Holme 1984: Figure 6.17; Brown et al., 2002: Figures 4 and 6; Eleftheriou and Moore, 2005; Figure 5.14. Unlike the other grabs considered here, the Hamon grab has a single large scoop-like bucket. This is fixed to the end of a long arm and mounted in a large pyramidal frame with a square/rectangular base (Figure 17–8 and Figure 17–9). The pivot point lies a little way along the arm and not at the top of the bucket, as with the preceding grabs. A hook-like trigger nearer the free end of the arm engages with the frame base and holds the grab open under tension from the warp. On arrival at the seabed, the tension is released, and the hook disengages. On hauling, the warp runs through two pulleys, causing the lifting arm to rotate through 90°, driving the sampling bucket (scoop) through the sediment and closing it against a rubber-covered stop plate. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 158 Figure 17–8. Hamon grab being deployed Photo: Craig Brown/Cefas. Figure 17–9. Hamon grab schematic Plate taken from Eleftheriou and Holme (1984). The grab was originally designed for collecting material from hard, coarse substrata off the Dutch 2 coast. The original design was very large and took samples covering an area up to about 0.29 m . 2 Subsequent use and modification have led to the production of the more manageable 0.1 m version, which is the officially recommended sampler for UK studies in areas of aggregate extraction (Brown et al., 2002). The adoption of the grab in the UK in the early 1990s (e.g. Kenny and Rees 1996) led to a number of design modifications by Cefas (Kenny, pers. comm., Limpenny, pers. comm.). Apart from the introduction and testing of the smaller model, adjustments concerning the weighting of the grab and the use of a longer toothed bar on the underside, have led to a grab with better sample reproducibility and efficiency. The grab works very well on most coarse grounds, although cobbles are understandably difficult to sample (Marine Ecological Surveys, per.s comm.; pers. obs.). 2 The 0.1 m Hamon grab weighs about 300 kg and the addition of weight to the frame (Figure 17–8) can take this to around 600 kg. Insufficient weighting can cause the grab to ‘walk’, sliding the frame along the sediment in the opposite direction to the movement of the sampling bucket. Some foreshortening of the sample area will occur if the pivot of the grab is raised by ‘warp-heave’. 2 In normal operation, the 0.1m grab provides samples of 10 – 12 litres (Limpenny, pers. comm). The stop plate can be fitted with an inspection hatch, but accessing the sample is difficult. The motion of the scoop through the sediment tends to mix the sample, so the Hamon grab is not used in studies that require the undisturbed surface of the sample to be viewed or sub-sampled. Failure to take a sample is unusual, compared to grabs designs that use two opposing jaws, but can be caused by stones being wedged between the scoop and stop-plate (Limpenny, pers. comm., pers. obs.). 5 – Review of Existing Standards and Protocols There are a number of publications on standards and protocols relating to the use of grabs to acquire good benthic data. This review includes information from UK, Europe and North American sources. Some of the standards and protocols were developed for use in specific types of study (e.g. sewage dumping grounds, aggregate extraction), others relate to specific types of grab, and some are based on personal experience. Some have more universal application than others, with a few reflecting relatively local ‘historical’ procedures that are to be maintained to ensure long-term compatibility of temporal datasets. Whatever the rationale behind the different accounts, all include useful information. This review focuses primarily on faunal sampling using grab devices. For a consideration of sediment sampling using grabs, the reader is directed to the separate review of particle size analysis (granulometry) by Passchier (this volume). Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 159 5.1 – Data Acquisition The following text provides short summaries of a number of notable publications. They are arranged chronologically, European studies followed by those from North America. European sources ICES soft-bottom macrofaunal sampling publication Sometimes referred to as the ‘ICES Green Book’. Rumohr (1990) produced a guide to the collection and treatment of benthic samples. A new edition of this valuable publication was produced more recently (Rumohr, 1999) to incorporate the results of ICES/HELCOM quality assurance workshops (ICES, 1996) into the recommendations. It considers, inter alia, sampling strategy and equipment, sample processing and quality assurance. For grabs, Rumohr noted that there was no unequivocal evidence that any one grab performs consistently better than its counterparts, in all conditions. He concluded that the use of all standard designs (Van Veen, Smith-McIntyre and Day grabs) be continued. The ‘Yellow Book’ So-called because of its yellow cover, this work (Rees et al., 1990) was produced as a guide for those involved in sampling UK sewage sludge dumping sites. It is one of the first UK guides to comprehensively examine all aspects relevant to a specific macrofaunal monitoring programme, from the initial planning to archiving and publication. A useful flow diagram presents the decisions that need to be made regarding any benthic sampling programme, and recommendations made concerning the specification of grab samplers and the processing of the resulting samples. JAMP benthic sampling guidelines The OSPAR Joint Assessment and Monitoring Programme (JAMP) sets out the rationale, objectives for quantitative benthic sampling, and the sampling, analysis, data and quality assurance requirements (OSPAR 1997). Technical Annex 2 concerns the soft-bottom macrofauna. Some of the advice is specific (field data recording), but most is general and reference is made to other publications for the detail (ICES 1994, Rees et al., 1991, Rumohr 1990). SAC Marine Monitoring Handbook Thomas (2001) provides a ‘Procedural Guideline’ for quantitative sampling of sublittoral sediment biotopes and species using grab samplers. The work is geared towards standardising practices for studies relating to Special Areas of Conservation (SACs), and would be relevant to seabed habitat mapping. The advice given is clear and practical, with the pros and cons of different equipment, their use in the field, shipboard sample processing, laboratory work, quality control and safety all being discussed. The appendices include equipment checklists, safe working practices on boats, plus deployment procedures for the Day and Hamon grabs. CEFAS marine aggregate area sampling guidelines Although produced to help standardise benthic work in aggregate-related work, this comprehensive work (Boyd 2002) presents clear, well-laid out information relevant to other studies — including seabed mapping. The chapters cover all aspects of survey from planning through to the reporting stage. Techniques covered include grab sampling and sedimentology. A useful table summarising the specifications, and pros and cons of the different benthic samplers is presented. A standard operating procedure (SOP) for the Hamon grab is presented and covers everything connected with using this apparatus and processing samples obtained from it. Standard Operating Procedures (SOPs) Review Cooper and Rees (2002) examined 23 standard operating procedures (SOPs) submitted by 6 organisations participating the UK’s National Marine Biological Analytical Quality Control scheme (NMBAQC). This is an excellent publication covering all aspects of field and laboratory work, including grab sampling and processing. The review assesses all the procedures and gives a set of recommendations for raising quality standards. The Green Book Produced by the UK Marine Pollution Monitoring and Management Group (2003), the “Green Book” provides procedural guidelines for the collection, processing and analysis of a variety of environmental samples. The guidelines are designed to enable the fulfilment of the UK’s commitments to a number Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 160 of local, national (e.g. National Marine Monitoring Programme) and international (e.g. European Community, OSPAR, JAMP) programmes and agreements. Appendix 1 (6 pages) concerns quantitative sublittoral macrobenthic sampling. It details the requirements relative to sampling 2 strategy, precision of site positioning, type of grab (0.1 m Day or Van Veen), collection procedures, sample processing and data analysis. Humber Benthic Field Methods Workshop Proceedings In 1997, an important workshop on benthic field methods took place at Hull University (Proudfoot et al., 2003). This workshop – attended by representatives from 11 benthic laboratories – yielded a wealth of information on the all aspects of benthic work and a best practice protocol for the use of 0.1 2 m grabs. Procedures in other available guidelines – the “Yellow Book” (Rees et al., 1990) and ICES ‘Green Book’ (Rumohr 1990) – were reviewed and commented upon prior to the production of a set of proposed UKNMBAQC field methods. The appendices give detailed standard operating procedures for the Day, Hamon and Van Veen grabs. ICES/OSPAR Quality Assurance publication Aimed at ensuring quality standards in a variety of biological studies, this important publication (Rees, 2004) also contains a wealth of information on macrozoobenthos, sampling design, and field surveys, and laboratory work. Annex 5 details good practice in relation to benthic macrofaunal sampling and analysis and contains some pertinent points aimed at achieving consistency: e.g. Sampling devices (grabs, corers, etc.) must be used on a long-term basis; gear changes have to be accompanied by intercalibration and a period of parallel sampling. Methods for the Study of Marine Benthos Eleftheriou and Moore (2005), in the third edition of this standard work, detail the current situation concerning macrofaunal sampling by trawl, sledge, corer and other methods. Previous editions are Holme and McIntyre (1971) and Holme and McIntyre (1984). The book covers all aspects of benthic sampling and is a core publication for all involved in seabed study. Three key chapters relating to benthic grab sampling are: Chapter 1 - Design and analysis in benthic surveys (Underwood and Chapman); Chapter 5 - Macrofaunal techniques (Eleftheriou and Moore) and Chapter 7 - Deep-sea benthic sampling (Gage and Bett). International Standard ISO/DIS 16665 These guidelines for quantitative sampling of the benthos were developed over a number of years by the European Committee for Standardization (CEN) ISO Technical Committee 147/5 and were recently published (ISO 16665: 2005). The guidelines have been formulated with reference to many of the foregoing publications (e.g. Eleftheriou and Holme 1984; Rees et al., 1990; Rumohr 1999; Proudfoot et al., 2003) and coverage is therefore comprehensive. North American sources Puget Sound benthic sampling protocols These protocols (Tetra Tech, 1987) originated from a 1985 workshop and a series of reviews written by representatives of most of the organisations involved in Puget Sound benthic work. The prime aim of the publication was to standardise the methodology, creating compatible data sets and enabling the creation of an inclusive Puget Sound database. The publication considered nine elements that needed to be addressed for standardisation. These are listed below, along with the most prevalent choice among the organisations involved in the Sound: • Sampler modified Van Veen • Sample area 0.1 m2 • Replication 4 or 5 samples per station • Sieve mesh 1.0 mm • Sieving location on vessel • Relaxant use no • Stain use Rose Bengal • Taxonomic level species • Sampling season variable If a 0.5 mm sieve was used it was recommended that 1.0 and 0.5 mm sieve fractions be obtained. Further, should identifications only be carried out at a higher-level (e.g. class), material should be Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 161 archived to allow future species-level analysis. Minimal acceptable sampler penetration was graded according to substrate, ranging from 4-5 cm in coarse sand to !10 cm in muddy sediment. Macrofaunal sample processing is described in precise detail and templates are provided for sampling logs, infaunal ‘chain of custody’ sheets, and tracking the progress of infaunal laboratory processing. Requirements for quality control of sorting and identification are also given. New Brunswick Marine and Estuarine Biodiversity Monitoring Protocols Pohle and Thomas (1997) describe these protocols in the form of a comprehensive and wellreferenced review. Extensive information is supplied on all aspects of benthic work from sampling to data analysis. Later, Pohle (1999) gave detailed protocols for sampling the infauna and other faunal groups. Experts and student volunteers were used to test the protocols at specific sites. Protocols were written in a logical step-by-step manner taking the participants through all the stages from the personnel and equipment required through to specimen identification and verification. A 0.04m2 Ponar grab was used. US EPA benthic manuals There are a number of publications by the US Environmental Protection Agency that provide information concerning benthic sampling. Section 3 in Strobel et al. (1995) gives an account of laboratory methods for macrobenthic community assessment. By contrast, section 6 of the Field Operations Manual of the National Coastal Assessment (Strobel and Heitmuller, 2001) provides a thorough account of field sediment sampling. A 2 step-by-step protocol is given for the operation of a 0.04 m Young grab and processing for benthic fauna, sediments and contaminants. Southern California Bight Operations Manual The field operations manual of the Southern California Bight Field Methods Committee (2002) was produced as a guidance document aimed at achieving consistency in the methods and procedures used by over 60 organisations. The manual gives concise, but pertinent advice on all aspects of field 2 sampling for a range of techniques including grabs and trawls. A 0.1 m Van Veen grab and a tandem Van Veen (both chain rigged) are described as is the operation of the grabs and the processing of samples. An interesting difference from other manuals is the recommendation that relaxants (Epsom Salts or propylene phenoxytol) be routinely used. From personal experience, such treatment must be carried out with great caution. Different animals react in different ways to such chemicals and all will have different response times. The danger is that some will be over-relaxed and in poor condition when finally fixed. A safer alternative is to selectively relax particularly fragile forms, though this is only possible for larger animals removed during the sieving process. Chesapeake Bay Quality Assurance Project Plan This publication (Versar, 2002) describes the standard operating procedures relevant to all aspects of long-term monitoring in Chesapeake Bay since 1984. High quality data throughout was highlighted to ensure the required standards of accuracy were achieved. Much of the text is written relative to the responsibilities of the various staff involved. 5.2 – Quality Control and Quality Assurance Quality control and assurance are important elements in acquiring faunal data from grab samples. In their thorough review of Standard Operating Procedures, Cooper and Rees (2002) recommended that: … individual laboratories examine each stage of a procedure and, where there is the potential for variation in the quality of output, decide upon acceptable boundaries. Where a specific level of accuracy or precision is required then this should be stated and a Quality Control procedure established to ensure that standards are met. At the very least, the quality control procedure should document the standard attained. By highlighting the quality of the data it is possible to ensure that it is not used in an inappropriate way. Rees (2004) sets out the guidelines for quality assurance recommended by the ICES/OSPAR Steering Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 162 Group on Quality Assurance of Biological Measurements in the Northeast Atlantic (SGQAE), which covers sampling design, field surveys, and laboratory work. Annex 5 details good practice in relation to benthic macrofaunal sampling and analysis. The accurate identification of specimens is crucial for any analysis to be valid. The taxonomic competence of personnel is not usually assured by a recognised qualification, rather through training workshops. In the UK, approximately 30 laboratories participate in the National Marine Biological Analytical Quality Control Scheme (NMBAQC), which formally tests faunal identification of batches of material circulated to the laboratories. Within a laboratory it is common practice for a proportion of samples processed by an individual to be cross-checked by a fellow worker to ensure that all fauna have been picked from the sample and the taxa correctly identified. Many laboratories also seek to provide quality assurance by sending a proportion of samples for external checking and verification. They may also send entire reference/validation collections for external scrutiny. 5.3 – Data Processing The raw data obtained by acoustic sampling techniques requires extensive processing before it can be analysed and interpreted, but this is not the case for the raw faunal data obtained by grab sampling. Data are usually provided in the form of species-by-sample tables, which record the abundance and biomass of the observed taxa. Prior to analysis, data processing is usually limited to two specific procedures aimed at standardising the data: 5.4 – Standardisation of the area sampled When working with infaunal data (i.e. that derived from grab or core samples), it is convention to express faunal density (or biomass) in terms of numbers (or weight) per unit area of the seabed (not per unit volume of sediment). The area of seabed sampled by any grabbing device should be known. 2 Most grabs in common usage (see the examples above) sample an area of 0.1m , so it is a simple calculation to convert the data to numbers (or biomass) per square metre. 5.5 – Standardising taxonomy, including data truncation The field of taxonomy is continually advancing and there can be many changes made to taxonomic nomenclature. Consequently, it is common for a study to use a recognised species checklist drawn from the taxonomic literature at a set point in time (e.g. Howson and Picton, 1997). It is important that all studies declare which taxonomic literature has been used in making faunal identifications. When faunal data are brought together from disperate studies there can be a requirement to harmonise the taxonomy across the different data sets. Where different studies have identified fauna to different taxonomic levels, a process of data truncation may be required to harmonise the data prior to analysis. For instance, if one study has identified taxa to species level, but another has only identified taxa to genus level, then the data from the first study must be truncated to genus level before the two data sets can be pooled. Often, a taxonomic coding system will be used to aid data manipulation and analysis, giving each taxon a specific number or letter code. Some systems are hierarchical, giving related taxa the same root code (e.g. NODC Taxomomic Codes). Such systems frequently undergo revision to keep up to date with revisions in taxonomy, and this can lead to incompatibilities between different versions of the same coding system. An alternative approach is to use a non-hierarchical structure, assigning each taxon a permanent serial number that does not change in line with any taxonomic revisions. This is the basis of the Taxonomic Serial Number (TSN) used by the Integrated Taxonomic Information System (ITIS). A variety of common coding systems have been considered by Vize and Coggan (2006) in their review of trawling and dredging techniques (this publication). 6 – Provenance and Current usage Quantitative grab sampling as we know it today can be traced back to the pioneering work of the Danish C. G. J. Petersen in the early 1900s (Petersen, 1914, 1918; Petersen and Boysen Jensen, 1911). The early Petersen grab was soon superseded by the Van Veen grab, which was shown to be a superior device and is still in common use. (In 2002, the American Society for Testing and Materials withdrew its Standard Practice Guide for the Petersen grab.) The Smith-McIntye grab became popular Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 163 in the 1950s and ‘60s and is used worldwide to the present day. However, in northern Europe, the similarly framed Day grab (Day 1978) is preferred on account of its simpler mechanism and greater reliability. In recent years, the commercial extraction of gravel from the seabed has led to an increased interest in the fauna supported by harder unconsolidated sediments, to which the twinbucket van Veen, Smith-McIntyre and Day grabs are not well suited. Consequently, the single bucket Hamon grab has recently come to prominence and is widely used for sampling gravel and sand substrates. As pointed out by Eleftheriou and McIntyre (2005: preface) there has been little change in macrofaunal sampling methodology in 20 years. The L-framed Van Veen described herein (and Mackie, in prep.) is arguably the only recent development of note. 7 – Conclusions and Recommendations 7.1 – The Sampler The actual grab design employed can vary considerably between countries, particularly between Europe and the USA. This is likely to remain the case for historical reasons – a particular grab is often specified in long-term monitoring studies for reasons of consistency as much as anything. For habitat mapping studies it would be beneficial to limit the variety of grabs used as this would promote consistency across studies and minimise data artefacts that may be attributable to the use of different grab designs. All of the most commonly used European samplers have strengths and weaknesses, and each can excel in a particular application. Consequently it would be inappropriate to recommend any one type of grab as a ‘standard’ for habitat mapping studies. Instead, guidance should be given on the selection of grabs, so they are matched to the type of substrate to be sampled. The van Veen grab is arguably the most versatile, being suitable for use on all substrates except hardpacked gravels and cobbles, and from all sizes of vessel. However, the lack of a supporting frame makes it more prone to failure than other designs. The Day grab is compact and has a similar versatility regarding ease of deployment, though it has more limitations on softer and harder sediments. The Hamon grab is suitable for both sand and gravel substrates, but should not be used to sample muds, on account of its great weight, which can cause the grab to sink deeply into soft muds. When setting out to sample unknown grounds, it would be advisable to take a Hamon grab and one of the twin-bucket designs (van Veen, Day or Smith McIntyre grab). One problem with assessing grabs is the variety evident within each design. Different examples of the same grab type may differ in their characteristics, so it is recommended that more information (metadata) is routinely recorded about the design and rigging of the grab (e.g. area of seabed sampled, shape of buckets, maximum possible penetration, modifications to the frame, meshes used on topplates, rigging and the amount of weight used). It is also recommended that metadata relating to each sample is fully recorded, including position co-ordinates, time and depth of sampling, sea conditions, the volume of the sample and its nature (e.g. sediment type from visual inspection). In addition, each sample should be photographed prior to processing to provide a permanent record of the appearance of the sample. Although habitat mapping is less reliant on fully quantitative data than monitoring studies, these types of metadata are important in assessing whether apparent faunal differences between samples (and /or habitats) are real or simply reflect the use of different sampling devices. There is still a need for cross-comparisons within and between grab types concerning bite profile, volume of sediment collected and faunal capture efficiency. Such a comparison may be beyond the scope of the present project. 7.2 – Standards and Protocols This review has shown that there are many existing standards and protocols relating to the use of grabs and the processing of the resulting samples. Several are designed for environmental monitoring or surveillance reasons – for example concerning sewage sludge dumping, aggregate extraction or point source discharges. Some are retained for historical reasons; that is providing consistency in sampling over long time periods, and may be local, national or international in their application. Most contain some information that is, or should be, universal. Referring future habitat mapping studies to the plethora of available documents would likely lead to a great deal of confusion and ambiguity in Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 164 precisely how grab sampling techniques should be applied for habitat mapping. recommended that specific guidance should be developed by the MESH project. Therefore, it is Existing protocols detail a variety of sampling strategies. As each is developed for a particular purpose, it is not surprising that they differ in their recommendations relating to the number of sampling stations and sample replication. Consequently, this is another area of guidance that will need to be provided specific to habitat mapping studies. Whereas protocols for environmental studies are designed to obtain numerical data to a desired statistical precision, this is not a fundamental requirement for benthic habitat mapping, where the major interest lies in describing species assemblages over a particular spatial area. While it is certainly feasible to produce detailed fine resolution maps of benthic faunal assemblages from intense surveys involving many stations with a high number of replicates, this is neither pragmatic nor cost-efficient. When applied to habitat mapping, grab sampling will most frequently be employed as a method of ‘ground- truthing’ acoustic surveys, sampling within different seabed facies (distinct sediments and bedforms) to ascertain the nature and variety of faunal assemblages that they support. Given limited resources (time and money) it would be preferable to increase the number of sampling points (‘stations’) and reduce the number of replicates per station. Specific guidance will need to be developed on sampling strategies and the minimum requirement for ground-truth sampling. Standards and protocols relating to sample processing seem to have become more firmly established in the past decade, in response to the growing emphasis on monitoring studies where consistency in methodology is of utmost importance. Such protocols would appear to be suitable for use in habitat mapping studies, without major revision. For the purpose of standardising procedures across future mapping studies, it is recommended that existing protocols should be drawn together in a single guidance document. Acknowledgements Many thanks for all the help provided by Ivor Rees, Teresa Darbyshire and Kate Mortimer. We are grateful to Gerald J. Kahl (Kahl Scientific Instrument Corp., California) for permission to use images of the Smith-McIntyre grabs (Figure 6). Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 165 REFERENCES Andersin, A.-B. and Sandler, H. (1981). Comparison of the sampling efficiency of two van Veen grabs. Finnish Marine Research 248: 137-142. Armonies, W. (2000). On the spatial scale needed for benthos community monitoring in the coastal North Sea. Journal of Sea Research 43: 121-133. Bachelet, G. (1990). The choice of a sieving mesh size in the quantitative assessment of marine macrobenthos: a necessary compromise between aims and constraints. Marine Environmental Research. 30: 21-35 Birkett, L. (1958). A basis for comparing grabs. Journal du Conseil Permanent International pour l'Exploration de la Mer 23: 202-207. Boyd, S. E. (2002). Guidelines for the conduct of benthic studies at aggregate dredging sites. Department for Transport, Local Government and the Regions, London (Report produced by the Centre for Environment, Fisheries and Aquaculture Science). 117 pp. [www.cefas.co.uk/publications/files/02dpl001.pdf]. Brown, C. J., Limpenny, D. S. and Meadows, W. (2002). The conduct of benthic surveys at aggregate extraction sites. In Guidelines for the conduct of benthic studies at aggregate dredging sites (ed. S. E. Boyd): 21-41 and 102-112 (Annex II). Department for Transport, Local Government and the Regions, London (Report produced by the Centre for Environment, Fisheries and Aquaculture Science). Cooper, K. M. and Rees, H. L. (2002). National Marine Biological Analytical Quality Control Scheme (NMBAQC). Review of Standard Operating Procedures (SOPs). Science Series. Aquatic Environment Protection: Analytical Methods., CEFAS Lowestoft, (13): 57 pp. Cuff, W. (1980). Comment on optimal survey design: reply. Journal of the Fisheries Research Board of Canada 37(2): 297. Cuff, W. and Coleman, N. (1979). Optimal survey design: lessons from a stratified random sample of macrobenthos. Journal of the Fisheries Research Board of Canada 36(4): 351-361. Day, G. F. (1978). The Day grab - a simple sea-bed sampler. Oceanographic Sciences 52 Report of the Institute of Eagle, R. A., Norton, M. G., Nunny, R. S. and Rolfe, M. S. (1979). The field assessment of effects of dumping wastes at sea. 2. Methods. Fisheries Research Technical Report, Ministry of Agriculture, Fisheries and Food Directorate of Fisheries Research 47: 1-24. Eleftheriou, A. and Holme, N. A. (1984). Macrofauna techniques. In Methods for the study of marine benthos. IBP Handbook 16 (second edition) (ed. N. A. Holme and A. D. Mcintyre): 140-216. Blackwell Scientific Publications, Oxford. Eleftheriou, A. and McIntyre, A. D. (2005). Methods for the study of marine benthos (third edition). Blackwell Scientific Publications, Oxford. Eleftheriou, A. and Moore, D. C. (2005). Macrofauna techniques. In Methods for the study of marine benthos (third edition) (ed. A. Eleftheriou and A. D. Mcintyre): 160-228. Blackwell Scientific Publications, Oxford. 418 pp. Elliott, J. M., Tullett, P. A. and Elliott, J. A. (1993). A new bibliography of samplers for freshwater benthic invertebrates. Freshwater Biological Association Occasional Paper 30: 1-92. Farris, R. A. and Crezee, M. (1976). An improved Reineck box corer for sampling coarse sand. Internationale Revue der Gesamten Hydrobiologie 61(5): 703-705. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 166 Gage, J. D.. and Bett, B. J. (2005). Deep-sea benthic sampling. In Methods for the study of marine benthos (third edition) (ed. A. Eleftheriou and A. D. Mcintyre): 273-329. Blackwell Scientific Publications, Oxford. Green, R. H. (1979). Sampling design and statistical methods for environmental biologists. John Wiley and Sons, New York. 257 pp. Green, R. H. (1980). Comment on optimal survey design. Journal of the Fisheries Research Board of Canada 37(2): 296. Holme, N. A. (1964). Methods of sampling the benthos. Advances in Marine Biology 2: 171-260. Holme, N. A. (1971). Macrofauna sampling. In Methods for the study of marine benthos. IBP Handbook 16 (ed. N. A. Holme and A. D. McIntyre): 80-130. Blackwell Scientific Publications, Oxford. Holme, N. A. and McIntyre, A. D. (1971). Methods for the study of marine benthos. IBP Handbook 16. Blackwell Scientific Publications, Oxford. Holme, N. A. and McIntyre, A. D. (1984). Methods for the study of marine benthos. IBP Handbook 16 (second edition). Blackwell Scientific Publications, Oxford. 387 pp. Hopkins, T. L. (1964). A survey of marine bottom samplers. In Progress in Oceanography (ed. N. Sears): 215-256. Pergamon Press. Howson, C. M. and Picton, B. E. (eds) (1997). The Species Directory of the Marine Fauna and Flora of the British Isles and Surrounding Seas. Belfast and Ross-on-Wye: Ulster Museum and the Marine Conservation Society. ICES (1994). Report of the ICES/HELCOM workshop on quality assurance of benthic measurements in the Baltic Sea. ICES C.M. 1994/E: 10 ICES (1996. ICES/HELCOM workshop on quality assurance and intercomparisons of benthos methods in the Baltic Sea. ICES C.M. 1996/E: 2. ISO 16665: (2005). Guidelines for quantitative sampling and sample processing of marine soft-bottom macrofauna. 30 pp. [www.iso.ch/iso/en/CatalogueDetailPage.CatalogueDetail?CSNUMBER=32348andscopelist=] Kenny, A. J. and Rees, H. L. (1996). The effects of marine gravel extraction on the macrobenthos: results 2 years post-dredging. Marine Pollution Bulletin 32(8-9): 615-622. Kingston, P. F. (1988). Limitations on offshore environmental monitoring imposed by sea bed sampler design. Advances in Underwater Technology, Ocean Science and Offshore Engineering 16: 273-281. Kingston, P. F. (2001). Grabs for shelf benthic sampling. In Encyclopedia of Ocean Science (ed. J. H. Steele, S. A. Thorpe and K. K. Turekian): 1169-1177. Academic Press, Kingston, P. F. and Riddle, M. J. 1989. Cost effectiveness of benthic faunal monitoring. Marine Pollution Bulletin 20: 490-496. Mackie, A. S. Y. (1981). Marine benthic macrofaunal sampling. Institute of Offshore Engineering, Heriot-Watt University, Edinburgh. 28 pp. Mackie, A. S. Y. in prep. A new modified Van Veen grab for more efficient operation and safer handling. Mackie, A. S. Y. and Darbyshire, T. (2001). Study area and benthic sampling. In Benthic biodiversity in the southern Irish Sea 2. The South-West Irish Sea Survey (ed. J. G. Wilson, A. S. Y. Mackie, Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 167 B. D. S. O'Connor, E. I. S. Rees and T. Darbyshire). Studies in Marine Biodiversity and Systematics from the National Museum of Wales. BIOMÔR Reports 2, pp. 33-47 and 105-120. Mackie, A. S. Y., Oliver, P. G. and Rees, E. I. S. (1995). Benthic biodiversity in the southern Irish Sea. Studies in Marine Biodiversity and Systematics from the National Museum of Wales. BIOMÔR Reports 1: i-viii + 263 pp. Marine Pollution Monitoring and Management Group (2003). UK National Marine Monitoring Programme Green Book. Marine Pollution Monitoring and Management Group [www.marlab.ac.uk/FRS.Web/Delivery/Information_resources/information_resources_view_documents .aspx?resourceId=18479] McIntyre, A. D. (1969). Ecology of marine meiobenthos. Biological Reviews (and Biological Proceedings) of the Cambridge Philosophical Society 44: 245-290. McIntyre, A. D. and Warwick, R. M. (1984). Meiofauna techniques. In Methods for the study of marine benthos. IBP Handbook 16 (second edition) (ed. N. A. Holme and A. D. McIntyre): 217244. Blackwell Scientific Publications, Oxford. Oele, E. (1978). Sand and gravel from shallow seas. Geologie en Mijnbouw 57: 45-54. Orton, J. H. (1925). On the efficiency of the Petersen grab. Nature, London 115: 156-157. OSPAR. (1997) JAMP (Joint Assessment and Monitoring Programme). Eutrophication Monitoring Guidelines: Benthos. (version 9.6.97). OSPAR Commission, London. 12 pp [www.ospar.org] Passchier, S. (2006). Review of particle size analysis (granulometry) of sediment samples. This publiction. Petersen, C. G. J. (1914). Valuation of the sea II. The animal communities of the sea-bottom and their importance for marine zoogeography. Report of the Danish Biological Station to the Board of Agriculture (Ministry of Fisheries) Copenhagen 21: 1-44 + 6 pls., 3 charts. Petersen, C. G. J. (1918). The sea-bottom and its production of fish-food. A survey of the work done in connection with the valuation of the Danish waters from 1883-1917. Report of the Danish Biological Station to the Board of Agriculture (Ministry of Fisheries) Copenhagen 25: 1-62 + 10 pls., 1 chart. Petersen, C. G. J. and Boysen Jensen, P. (1911). Valuation of the sea I. Animal life of the seabottom, its food and quantity (quantitative studies). Report of the Danish Biological Station to the Board of Agriculture (Ministry of Fisheries) Copenhagen 20: 1-81 + 6 pls., 6 tables, 3 charts. Pohle, G. W. (1999). Testing and implementation of EMAN biodiversity monitoring protocols for marine ecosystems and examples of site specific protocols. Final report to the Ecological Monitoring Coordinating Office, Canada. 37 pp. (25 pp. + appendices) [www.cciw.ca/emantemp/research/protocols/99_marine/intro.html]. Pohle, G. W. and Thomas, M. L. H. (1997). Monitoring protocol for marine benthos: intertidal and subtidal macrofauna. Ecological Monitoring and Assessment Network, Environment Canada. 32 pp. [www.eman-rese.ca/eman/ecotools/protocols/marine/benthics/intro.html]. Powers, C. F. and Robertson, A. (1967). Design and evaluation of an all-purpose benthos sampler. Great Lakes Research Division, Special Report 30: 126-131. Proctor, N. L., Elliott, M. and Quintino, V. (2003) Field and laboratory methods for analysing Marine and Estuarine Benthos and Fishes. Environment Agency Benthos Workshop - May 2003. Hosted by the Institute of Estuarine and Coastal Studies (IECS): Events and Conferences, [http://www.hull.ac.uk/iecs/]. Proudfoot, R. K., Elliott, M., Dyer, M. F., Barnett, B. E., Allen, J. H., Proctor, N. L., Cutts, N. D., Nikitik, C., Turner, G., Breen, J., Hemmingway, K. L. and Mackie, T. (2003). Proceedings of the Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 168 Humber Benthic Field Methods Workshop, Hull University 1997. Collection and processing of macrobenthic samples from soft sediments; a best practice review. Environmental Agency R and D Technical Report E1-135/TR: 1-viii, 1-128. Rees, E. I. S., Allen, P. L. and Coppock, J. (1994). Representative replication for sediment benthos monitoring: application of varied strategies in the Irish Sea. Porcupine Newsletter 5(9): 225-233. Rees, H. L. (ed.) (2004). Biological monitoring: General guidelines for quality assurance. ICES Techniques in Marine Environmental Sciences 32: 44 pp. Rees, H.L., Moore, D.C., Pearson, T.H., Elliot, M., Service, M., Pomfret, J. and Johnson, D., (1990). Procedures for the monitoring of marine benthic communities at UK sewage sludge disposal sites. Scottish Fisheries Information Pamphlet, 18: 78pp. Rees, H. L., Heip, C., Vinxcx, M. and Parker, M. M. (1991). Benthic communities: use in monitoring point source discharges. ICES Techniques in Marine Environmental Sciences 16: 1-70. Reineck, H. E. (1958). Kasterengreifer und Lotröhre 'Schnepfe', Gerate zur Entnahme ungestörter, orientierer Meeresgrundproben. Senckenbergiana Lethaea 39: 42-48. Reineck, H. E. (1963). Der Kasterengreifer. Natur und Museum 93: 102-108. Riddle, M. J. (1984). Offshore benthic monitoring strategies. Ph.d. Thesis, Heriot-Watt University, Edinburgh. 335 pp. Riddle, M. J. (1989a). Precision of the mean and the design of benthos sampling programmes: caution advised. Marine Biology, Berlin 103: 225-230. Riddle, M. J. (1989b) Bite profiles of some benthic grab samplers. Estuarine, Coastal and Shelf Science 29: 285-292. Rumohr, H. (1990.) Soft bottom macrofauna: Collection and treatment of samples. ICES Techniques in Marine Environmental Sciences 8: 1-18. Rumohr, H. (1999). Soft bottom macrofauna: Collection, treatment and quality assurance of samples. ICES Techniques in Marine Environmental Sciences 27: 1-19. Smith, W. and McIntyre, A. D. (1954). A spring-loaded bottom sampler. Journal of the Marine Biological Association of the United Kingdom 33: 257-264. Southern California Bight Field Methods Committee (2002). Field operations manual for marine watercolumn, benthic, and trawl monitoring in Southern California. Southern California Coastal Water Research Project Technical Report 359: 1-42 [www.sccwrp.org/tools/methods.htm]. Strobel, C. J. and Heitmuller, T. (2001). National Coastal Assessment: Field Operations Manual. U. S. Environmental Protection Agency, Office of Research and Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology Division, Gulf Breeze, FL. EPA 620/R01/003. 72 pp.. Strobel, C. J., Klemm, D. J., Lobring, L. B., Eichelberger, J. W., Alford-Stevens, A., Potter, B. B., Thomas, R. F., Lazorchak, J. M., Collins, G. B. and Graves, R. L. (1995). Environmental Monitoring and Assessment Program (EMAP) Laboratory methods manual. Estuaries Volume 1 – Biological and physical analyses. United States Environmental Protection Agency, Office of Research and Development, Narragansett, RI. EPA/620/R-95/008. 128 pp. Tetra Tech (1987). Recommended protocols for sampling and analyzing subtidal benthic macroinvertebrate assemblages in Puget Sound. 38 pp. [www.psat.wa.gov/Publications/protocols/protocol_pdfs/benthos.pdfande=8092]. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities 169 Thomas, N. S. (2001). Procedural Guideline No. 3-9: Quantitative sampling of sublittortal (sic) sediment biotopes and species using remote-operated grabs. In Marine monitoring handbook. March 2001 (ed. J. Davies, J. Baxter, M. Bradley, D. Connor, J. Khan, E. Murray, W. Sanderson, C. Turnbul and V. M.): 275-291 JNCC, Peterborough. [www.jncc.gov.uk] Thorson, G. (1957). Sampling the benthos. Memoirs. Geological Society of America 67: 61-73. Underwood, A. J. and Chapman, M. C.. (2005). Design and analysis in benthic surveys. In Methods for the study of marine benthos (third edition) (ed. A. Eleftheriou and A. D. Mcintyre): 1-42. Blackwell Scientific Publications, Oxford. Van der Meer, J. (1997). Sampling design of monitoring programmes for marine benthos: a comparison between the use of fixed versus randomly selected stations. Journal of Sea Research 37: 167-179. van Veen, J. (1933). Onderzoek naar het zandtransport von rivieren. De Ingenieur 27: 151-159. van Veen, J. (1936). Onderzoekingen in de Hoofden, in verband met de gesteldheid der Nederlandsche Kust. Dissertation, University of Leiden. Versar (2002). Chesapeake Bay Water Quality Monitoring Program: long-term benthic monitoring and assessment component, quality assurance project plan 2002-2003. Versar, Columbia, MD. 38 pp. [www.baybenthos.versar.com]. Vize, S. and Coggan, R. (2006). Review of standards and protocols for trawls and dredges. This publication. Wigley, R. L. (1967). Comparative efficiencies of Van Veen and Smith-McIntyre grab samples as revealed by motion pictures. Ecology 48: 168-169. Word, J. Q. (1976). Biological comparison of grab sampling devices. Southern California Coastal Water Research Project. Annual report 1976: 189-194. Review of standards and protocols for seabed habitat mapping – Video & Imagery facilities View publication stats 170