l. INTRODUCTION
1.1. These-comments arise from a request by the Director of South Wales Sea Fisheries Committee to provide an it~dependent evaluation of those aspects of the Gibb Wales (Aug 1997) Environment Statement for Proposed Dredging on the Helwick Bank that deal with studies of the benthic fauna and fisheries.
1.2. Although not having any direct experience of seabed sampling in the immediate vicinity of the Helwick Bank, I have had over 30 yeais experience of benthic sampling in somewhat similar tide swept parts of the Irish Sea and St Georges Channel (eg Mackie, Oliver and Rees, 1995). I also have experience of benthic sampling in areas licensed for aggregate dredging in Liverpool Bay and have been to sea on a trailer suction dredger offthe Humber specifically to observe the benthos and sand-eels brought up during commercial aggregate dredging. As a member of inter-departmental working groups I have been closely involved in advising on the protocols for monitoring the seabed in both sludge and dredge disposal grounds.
l .3 . In addition to .the Gibb Wales document, I had available to me as background, the HR Wallingford Report (EX 3546, Aug 1997) "Dredging on Helwick Bank; Study of Impacts on the Coastline; Formal Stage Report.

2. BENTHOS SAMPLING
2.1 In the Gibb Wales report the benthos sampling, specifically for this statement, is dealt with in section 5.4. This part amounts to only about 7 pages, mostly consisting of very widely spaced tables, out of a document running to 100+ pages. It was surprising to see more attention given to distant terrestrial SSSIs, where any connection with the project would be highly tenuous, than to the ecology of the seabed direetly to be impacted by this offshore marine project.
2.2 The report indicates that a sampling survey for the benthos and sediments was carned out in March 1997. No details are given of the actual dates, the sea conditions prevailing or the tidal states at the time. This last is particularly relevant to understanding the apparent ineffectiveness of the beam trawl sampling. There is also no mention in the report of what sort of vessel was used or how the gear was deployed (SWSFC indicated to me that some sampling might have been d'one from the trailer suction dredging vessel and some possibly from a boat belonging to Swansea University). I am not familiar with the dredging vessel operated by the applicant company, but, assuming it is typical of the smaller class of specialist aggregate dredgers, the deck layout and machinery was probably less than ideal for deploying light fishing equipment and grab~. In the absence of more specific information I also had to guess at possible operational scenarios to explain some odd features of the results. It would have been of interest to know whether special trips were made to collect the samples or whether the sampling was fitted into the tidally limited time available on one or more cargo dredging. trips, before after the low water period when

dredging took place. It wouId also be relevant to know whether the sampling at sea was done by the dredger's crew alone or whether a team of experienc~d benthic ecologists ~ were in Charge of the sampling and did the washing of the benthos samples at sea.
2.3 Table 5.1 of the Gibb document gives positions as Latitude and Longitude where the grabs were reported as having been taken. It is notable that alI the positions are given to 2 decimal places of a Minute of Latitude & Longitud'e, but strangely all 88 of the second decimal places are zeros. No details are given of the navigational electronics used to obtain these positions, so there is a suspicion that Table 5.1 conveys a spurious degree of accuracy. Differential GPS would have been needed to reliably achieve this Ievel of navigational accuracy. The report also indica,tes that at least four deployments were made of the Day grab at each station (3 benthos samples and 1 sedimer~t). Unless time was spent bringing the vessel back to the precise position before each Iowering of the grab, tidal drift would have spread the sampling out well beyond 0.01 Minutes of Longitude. If the grab was deployed on a rope operated from the drum end of the anchor windlass, the drift interval between replicates would have been even greater than on a research/survey vessel or trawler where the grab would more likely have been operated directly on a wire off the main drum of a winch. In the absence of sonic depth readings taken at the same time it is di~cult to be~ sure where, relative to the crest of the bank or the positions of major sand waves, the samples were actually taken. In areas such as this, there can be subtle but ecologically important differences between the sediment on the crests versus that in the troughs between the sand waves. Special care would have been needed in the sampling operation to account for this. The positions in Table 5. i appear more likely to be those set out in a set of pre-survey sampling instructions, than to be those actually Iogged at the time by the vessel's navigation officers or down-loaded Iater from some electronic track plotting or data recording system.
2.4 The Gibb report only says that the grab samples were wet sieved at sea using a lmm sieve. Linear banks similar to the Helwick tend naturally to have rather sparse b~enthic faunas, especially where there are obvious mobile sand waves and mega-ripples,.However Table 5.2 shows that only 10 animals were identified at the laboratory stage from an assumed 3 x 44 = 132 grab deplvyments, each grab being from O.lm2. Such an exceptionally Iow figure calls for a special explanation if it is to be believed. As not all the samples were reportedly taken on the actual bank this number of individual animals is even more surprisingly low. In trying to artive at an explanation for this, I note that all the species recorded in the list are those that would readily be noticed by inexperienced people looking at the contents remaining on a sieve after washing a sand sample through it with a hose on deck at sea. It is notable that not a single specimen was recorded of the smaller crustaceans such as Bathyporeia spp. or other amphipods that might have been expected to live in this mobile sand habitat. As there is no mention of the process of sorting the benthos in the laboratory from the bulk of dead shell and other ma,terial normally retained on I mm sieves, I would suspect that the fauna shown in Table 5.2 merely represents the obvious organisms that were picked offthe sieve at sea. If this was indeed the case then although it is possible for an experienaed benthic ecologist to guess at the general biotope types present, theprocedure adopted would have been at variance with nomally accepted protocols and not adequate to properly describe the ecoiogy of the area.
2.5 It is clear from Section 5.4.2 that even the authors of the Gibb report found the beam trawling ineffective and inadequate. Points of detail missing are whether the speed of tow given is speed over the ground (estimated by the navigation electronics or perhaps acoustic doppler) or speed of the vessel through the water. There is no iniormation on whether the light beam trawl was towed on a rope or a wire, which would have helped to hold it down. There is no mention of whether tickler chains were used with such gear. Although directions of tow are noted, this was not related to tidal currents at the time. There also appears to be confusion about the mesh of the net. The mesh size information in the report implies that the cod-end had a larger mesh than the main part of the net. This would be the reverse of the state in which small mesh nets are normally specified for sHrimp fishing by the net making and fishing gear supply companies. One might have expected mention of problems with the trawl sanding up, given the distances it was apparently towed across the crests of sand waves.
2.6. The list of species found in th~e trawls, as far as it goes, looks fairly nomal for such an area, though it does again appear to be limited to the eye-catching mobile species One would have expected to see some mention of very common hydroids such as Hydralmania falcata and bryozoans such a Flustra foliacea, even if it would not have been clear whether they were loose &agments or had actually been attached. One might also have expected such a net to have picked some dead shells with attached epifauna such as Pomatoceros spp. or small colonies ofAlcyonium digitatum. The presence ofPandadus montagui would perhaps suggest that not all the trawls were over very mobile sand without any clumps of sessile epifauna. As with the grabs the list given has the appearance of samples selectively picked over on deck by non-biologists, such as would have been the case if crew members on the dredger had been asked to take some samples for a biologist ashore to identify at a later stage. It is not clear whether the same person identified the beam trawl material and the grab material at much the same time. I noticed that the single species of mysid recorded in the trawls was different from that in the grabs. One might also have expected that at these depths, Crangon allmani might have been present as well as the common brown shrimp C. crangon.
2.7 I was surprised not to find any comparison in the main benthos section with previous studies of the benthos of the Bristol Channel and partieularly that by Tyler and Shackley ( 1980) The Tyler and Shackley paper is mentioned elsewhere in a catalogue of earlier reports but because of the way this report is organised the comparisons get lost.
2.8 The HR Wallingford report contains a very usefi~l map of seabed types based partly on side-scan surveys. Full use does not seem to have made of this in interpreting the benthos results in lhe Gibb report. Nor does it seem to have been used in planning the most effective stratified sampling of the dredging area and it's surroundings.

3. FISH AND FISHERIES
3.1 In ecological terms, sand eels are clearly the most important fish species using the Helwick Bank. Without much more data it is dif~cult to be clear how important this particular bank may be relative to other banks and sand wave fields in the same general part of the Bristol Channel, As the bank is a quasi-stable feature in longterm dynamic equilbium and is the most obvious feature in the area likely to have suitable sediments for sand-eels, it could be particularly important. It is acknowledged that assessing sand-eel stocks is rather difficult. To do this properly would probably require a combination of acoustic surveys and a lot of grab sampling of other sand features picked out by side-scan or acoustic ground discrimination methods. Account would need to be taken of the shoaling movements of the sand eels, sampling of the bed being done at times of day when the fish are most likely to have buried themselves in the sand. The present inadequate sampling merely supports the strong suspicion that the Helwick Bank is a particularly favoured habitat for sand-eels.
3.2 Sand-eels are particularly important in the diet of a number of fish species that have higher than average values both to the commercial fishery and as quarry species for sports fishermen. Obvious candidates as predators of sand-eels in this area off the Gower coast such as bass, turbot and brill highly valued both commercially and in the sports fisheries. Many other valued species (thornback ray, small-eyed ray, tope, sprdog, cod and whiting) can also be predators on sand eels. Mackerel and scad are attracted to concentrations of sand eels. Given the obvious identification of a sand-eel and fish food aspect to the likely dredging impacts one might have expected to see some attempt to examine in detail the importance of sand-eels in the diet of piscivorous fish caught in the area.
3.2 Although the by-catch of sand-eels during suction dredging is rnentioned in the Cribb report, it is implied that most of those fish surviving passage through the dredging pump will swim at the surface and go back over the side of the hopper with the wash-over water. In my experience they seem to swim to the deeper parts of the hopper and seem to enter the cargo of sand particularly as it is dried up at the end of the cargo loading period. This mirrors their behaviour when trapped in pools on sandy shores by the ebbing tide. Elsewhere, I have heard anecdotal reports of sand cargoes being rejected by customers because of the quantity of rotting fish contained and of concrete test blocks failing due to included fish. Advice given to me hy a very experienced master of an aggregate dredger was that to avoid contamination of the cargo with benthos and fish they tend to concentrate dredging along the same limited tracks within a wider licensed area. The ground is efiectively cleaned. It appears however that the Helwick operation is to be somewhat erratic, being governed by the vagiaries of the local market and the holding capacity of the shoreside storage area at Burry Port. Recolonisation by fish between dredging episodes on this mobile sediment bank seems more likely than where true fossil deposits of peri-glacial outwash sands and gravels are worked.
3.3 Anecdotal evidence only is given in the Gibb report on the importance of the area for both recreational and commercial fishing. In view of the greater potential for conflict on this topic than many of the other subjects covered in elsewhere in the report, one would have expected to see a real attempt to gather location specific effort and catch data. Had the environmental statement built upon a matrix table identifying the most significant interaction'problems to be addressed, it would probably have resulted in more attention being given to this aspect
3.4 There seems to be no mention of shellfish species. Given the previous conflicts elsewhere round Britain between aggregate dredging and migrations of edible crabs Carrcer pagurus, one would have expected at least some consideration of this. This species seems likely to be present of the slightly less mobile parts of the proposed area.

4. GENERAL ASPECTS
4.1 The Helwick Bank is a very significant feature of the morphology of the western part of the Bristol Channel off the Gower coast. In spite of some sand dredging having taken place in recent years this bank could be regarded as being in closer to a virgin natural condition than any of the other linear banks fizrther up the Bristol Channel. In ecological terms the bank is likely to be in a more natural condition and hence a better example of a habitat type included in the EU Habitats Directive, "banks covered at all times", than those mobile banks lumped into the Severn Estuary candidate Special Area of Conservation.
4.2 Given that the bank has been dredged in the past it would have been very useful for side-scan records to have been obtained to see whether the tracks made by trailer suction dredging persist for any length of time, or whether even on neap tides, sediment mobility obscures the tracks straight away. In preparing these comments I have assurned that the dredger will always trailer dredge while steaming slowly and will not anchor, so dredging pits in the bank. As the box which the applicants want to be licensed is only around 1.1 Nautical miles in the tidal axis direction and the shallow area to be dredged is only about 0.3 - 0.4 Nm wide it is to be expected that the vessel will be tempted stray over either end of the patch. It would be relevant to know if Crown Estates have required the particular vessel to be fitted with a black box recorder system.
4.3 Table 6.1 gives figures comparing Total Organic Carbon results in the present study with those found by Tyler and Shackley. The way the sites are numbered implies that the 11 earlier stations were in just the same positions as just the first 11 sites in the sequence done for this study. Some sort of statistical comparison between the two sets of samples would have been expected if temporal comparison is at all relevant.
4.4 Table 6.2 - Trace Metals. It is not clear whether like is being compared with like. Since most of the trace metals in any sediment are associated with fine fractions, many reports give figures for just the <64mu fraction, not total sediment. Had a matrix been used to identify the real issues of concern, the lack of trace metals in well washed sand would have soon eliminated the need for work on this aspect.

5. CONCLUSIONS
5.1 There are enough uncertainties in the ways the benthos and beam trawl surveys were apparently undertaken to cast considerable doubt on the validity of the data presented in the present Gibb report and it's interpretation in a wider context.
5.2 In spite of the extensive treatment given to many issues in the report, severai of the really key issues for fisheries interests are not adequately clarified.
5.3 A Welsh O~ce press release dated 1.8.96 announced a 2.5 year study of marine aggregate resources throughout the Bristol Channel and the environmental implications of working these. It seems premature to adjudicate on any increase in the aggregate exploitation rate on the Helwick Bank in isolation fiom the wider considerations.
5.4 The wider considerations must take account of the benthic ecology and fisheries aspects of the area as well as merely improving on models of sediment transport for reasons of coast protection engineering. It is not known to me whether the study announced by the Welsh Office in Aug~st 1996 includes any original work on seabed ecology, but for many of the reasons given above the sampling for the present Gibb Wales statement would be of limited use to anyone trying to take a broader view of the longterm sustainable management of biodiversity resources in the Bristol Channel.
6. REFERENCES
Gibb Wales. 1997. Environmental statement for proposed dredging on the Helwick Bank. Prepared for Llanelli Sand Dredging Ltd. August 1997 - Issue 1.
HR Wallingford. 1997. Dredging on Helwick Bank. Study of Impacts on the Coastline. Formal Stage Report. Report EX 3 546, August 1997.
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. BIOMOR Reports 1: 263pp.
Tyler, P.A. and Shackley, S.E. 1980. The Benthic Ecology of Linear Sandb~.nlcs: A modified Spisuls sub-community. In Industrial Embayments and their Environmental Problems.

FISH MAIN MENU