Extended sea pens are easily counted, either by first-hand observation or on videotape, except perhaps when populations are very dense, in which case a semi-quantitative abundance scale could be used (ie. occasional/frequent/abundant etc.). The periodic retraction of colonies of Virgularia and Pennatula into the substratum is a factor that could lead to errors in density estimates, as more pens will usually be present than are visible above the surface at any given time. Kinnear et al. (1996) noted the difficulty of accurately estimating Virgularia densities from towed video recordings. Birkeland (1974) found that the apparent density of Ptilosarcus guerneyi at his study sites varied from 0 - 5 colonies m^-2 to 10 - 30 m^-2, depending on the proportion of colonies extended. The average proportion of extended colonies was 26%. There was no obvious relationship between extension and tidal cycle, current strength or direction, turbidity, weather, season or time of day. Colonies were not synchronized in their behaviour, a feature also found by Hoare & Wilson (1977) for Virgularia mirabilis in Holyhead Harbour.

In the light of this behaviour pattern, counts of expanded sea pens should be regarded as giving minimum estimates of population density. If the retraction behaviour of colonies is not synchronized, this source of error should ‘average out’ when comparing sets of observations (ie. if, for example, 50% of colonies are extended at any given time, an observed density difference on successive observations is indicative of a population change, even if the absolute number of pens present is not known). The timing of expansion cycles, and the absolute number of colonies present, could be determined by observation of small areas of sea floor using a static video or time-lapse still camera deployed on the sea bottom for a period of a few days.

Burrow openings and sediment mounds can also be counted easily on dived transects or on good-quality video recordings. In the context of SAC monitoring it is recommended that counts should be made of broad categories of feature (eg. ‘Large/small mound’, ‘Large/small burrow opening’ etc.), rather than attempting to estimate precisely the population densities of the various species present (unless specialist help is available). There are numerous complicating factors involved in the latter exercise, and the time required to achieve it is far greater than for a basic count of surface features. The problems in converting from surface features to population densities include:

• Megafaunal mounds and burrow openings often show features diagnostic of particular species, but most bioturbated sediments contain many simple holes (large or small) that even an experienced observer will find difficult to identify. This ‘uncertainty factor’ is particularly acute in analyses of towed video recordings whose resolution may not be sufficient to show subtle identification features.
• The number of surface openings per individual burrow system is variable in many species (eg. Calocaris macandreae, Jaxea nocturna, Callianassa subterranea), so that converting from one parameter to the other is at best an approximation. In the north-eastern Irish Sea, megafaunal population density figures derived from surface mound and hole counts were found to underestimate the actual numbers extracted from box-core samples (Hughes & Atkinson, 1997). In the case of Callianassa subterranea a four-fold discrepancy was found, due partly to the presence of large numbers of small juveniles whose burrow openings were invisible at the scale of resolution of the towed video.
• At close range, a diver can quite easily distinguish the extent of an individual Nephrops burrow system by noting the relative orientations of the various openings. This is harder to achieve from a video recording, which may give only a fleeting view at a low level of resolution. The occurrence of vacant burrows, and others occupied by several animals (adult-juvenile complexes) creates problems in estimating animal densities from burrow densities. Again, if specialist help is not available, it is probably best to make a simple count of burrow openings rather than try to estimate the number of burrow systems present.

For these reasons, it is probably best to use the total number of mounds and burrow openings seen in a survey area as an indicator of the density and activity level of the megafaunal burrowing community. More precise, species-level density estimates can be made if sites are surveyed by experienced observers (especially if diving work is possible), but it is assumed that this will not always be possible. Specimens of the larger, more conspicuous megafauna (Nephrops norvegicus, Goneplax rhomboides, Cepola rubescens) may be seen above the sediment surface and should also be counted, but the number seen will usually be only be a small (and indeterminate) proportion of the local population.

Suggested basic categories of megafaunal surface features for monitoring purposes are listed in the table below, with guidance on the likely or potential creators of these. Size categories are very approximate, as all categories will show a continuous variation in size, with considerable overlap between the features made by different species. Marrs et al. (1996) give a more detailed guide to megafaunal surface features, incorporating the different size categories of these.

Atkinson (1989) and Marrs et al. (1996) found that counts of burrow openings and mounds made by divers agreed fairly well with those made over the same ground from towed video recordings. Diver observation allows a finer scale of resolution than is obtainable by video, so that more small surface features are likely to be recorded. Care should therefore be taken if figures obtained by the two methods are to be compared.

The larger epifaunal animals such as crabs, hermit crabs and starfish can be easily counted if desired. The burrowing anemones Cerianthus lloydii and (especially) Pachycerianthus multiplicatus should also be conspicuous when extended. Both species can withdraw into their tubes below the sediment surface, and population density estimates will therefore be subject to the same qualifications as for the sea pens discussed earlier.

The occurrence and extent of any surface patches of black, reduced sediment colonised by bacterial mats (Beggiatoa spp.) should be noted, as this will indicate a localized increase in sediment organic content. Such localized enrichment is a normal seasonal occurrence in many places, for example, in shallow sea lochs receiving a large input of loose seaweed, terrestrial leaf litter and settling phytoplankton, and so is not necessarily an indicator of adverse environmental changes. However, particular attention should be paid to the extent of these patches if the site is potentially influenced by local human input of organic matter (eg. a salmon farm or sewage outfall).

If sediment samples can be collected, it is possible to measure the organic content simply and easily by combustion. Small amounts of sediment (a few grammes) are freeze-dried, finely-ground, then heated to 500^oC for about 15 hours in a muffle furnace. After cooling, the samples are re-weighed, and the weight loss gives a measure of the organic content. The apparatus required to carry out this analysis (freeze-dryer, muffle furnace, accurate balance) is unlikely to be available on-site at a marine SAC, but will be standard equipment at any academic institution likely to be taking part in an SAC monitoring programme. Sediment samples can be stored frozen if immediate analysis is not possible.

The passage of a Nephrops trawl will generally leave conspicuous tracks on the sea bed, and the occurrence of these during a visual survey should be noted. A bottom trawl will leave two parallel linear tracks.

Next section                     References