Dominant taxa in the CFT.

Feeding strategies

Biological interactions and keystone species

By definition a 'faunal' turf must consist predominantly of animals, but as has been pointed out above the transition from the infralittoral is not clear cut, and the upper levels of the circalittoral will contain algae, the amount decreasing with depth. Crustose corraline algae are common (and usually unidentifiable in the field), and the other algae comprise a wide variety of small reds. Thus in a survey of the Dorset coastline (IOE Group, 1995) 16 species of red algae and 4 species of brown algae (Appendix) were collected from 'circalittoral bedrock'. This is probably representative of what is normally found.

There is a dominance of sessile species in the CFT communities, and so the animals would be expected to come predominantly from taxa which are basically sessile in habit. This is confirmed both by an analysis of the species composition on circalittoral bedrock in the Dorset survey cited above (IOE Group, 1995). An alternative approach, concentrating on the more prominent species, is to examine all of the 'characterising species' in CFT biotopes in the JNCC biotope classification (Connor et al., 1997) - these are listed in the appendix. The resulting breakdown is very similar to the above.

Dorset Survey Characterising

CFT Species

Sponges 28 23

Coelenterates 27 24

Tube worms 6 3

Barnacles 2 1

Molluscs 12 5

Bryozoans 25 12

Tunicates 11 14

Annelids 1 2

Crustacea 9 4

Echinoderms 3 11

Fish - 3

In the Dorset survey the sessile sponges, coelenterates, bryozoans and tunicates clearly dominate the community, making up more than 70% of the species. Of the remainder a number are also attached. Again the sessile taxa predominate among characterising CFT biotope species. The bryozoans are less well represented, since although common, individual species tend not to be conspicuous. The echinoderms make up a larger component - they tend to be very obvious elements in the community. However, it must be remembered that these breakdowns ignore the abundant cryptofauna which do not feature in any of the standard surveys.

The majority of the CFT organisms are filter feeders, depending on suspended material in the water column coming within their range. This suspended material may be living plankton, either plants in the form of phytoplankton or animals in the form of zooplankton, or it may be dead organic matter. The organic matter arises from various sources - such as by the breakdown of dead organisms, the shedding of fragments, the production of faeces, and the release of mucus. Since the majority of CFT communities occur in exposed or moderately exposed conditions the water movement will facilitate the supply of this suspended food. Phytoplankton tends to concentrate in the surface layers, but turbulent mixing will carry it down to the circalittoral zone. Much zooplankton makes diurnal vertical migrations, or remains at depth, making it available in the circalittoral zone. Particulate organic matter tends to settle to the sea bed, but the water movement will keep it in suspension. A number of the CFT taxa are clearly all or largely filter feeders - the sponges, tube worms, barnacles, bivalve molluscs, bryozoans, tunicates, and coelenterates. Certainly filter feeders predominate numerically, and they can be passive or active feeders.

Passive filter feeders depend entirely upon water movement to carry food particles to the filtering mechanism - hydroids and fan corals are of this type, as are most barnacles, and they would not be expected to flourish in extreme shelter. Active filter feeders, most of the other taxa (e.g. sponges, bryozoans, tunicates), create a water current to draw food into the collecting system. However, this current is in most cases quite weak and draws food from distances of a few centimetres at most. So whilst they can feed in static water, they nevertheless depend upon some water movements to replenish the local food resource. The coelenterates differ in feeding mode from the other groups above - they all depend upon a mixture of cilia, mucus and setal sieves to filter fine particles, and they can only filter fine particles, so they are obligate filter feeders. Coelenterates, however, have tentacles with stinging cells which they use to catch their prey, though some also employ mucus/ciliary mechanisms. Some feed only on small particles, and can be regarded therefore as functional filter feeders - this applies to hydroids, fan corals, and soft corals. The soft coral Alcyonium employs tentacular mechanisms to trap zooplankton, but at the same time filters phytoplankton with a mucus/ciliary mechanism (Roushdy & Hansen, 1961). Some anemones (e.g. Urticina) feed only on larger prey and are considered carnivores. However the plumose anemone (Metridium), whilst normally feeding on small particles, has retained the ability to capture larger prey.

A feature of filter feeders, particularly active ones, is their ability to modify the environment by reducing the concentration of suspended particles. This is probably only significant in semi-enclosed situations, but examples include the effects of mussel farming on the water clarity of fjord systems (Haamer, 1996), and of mussel populations in reclaiming disused docks (Wilkinson et al., 1996). In San Francisco Bay the bivalve population has the capacity to filter the volume of the bay daily, and is considered of far greater importance than the zooplankton in grazing down the phytoplankton (Cloern, 1982). Thus any change in the balance of filter feeders, both within or adjacent to CFT biotopes in enclosed situations, could affect water clarity and the supply of particulate food.

The conspicuous mobile taxa - decapod crustaceans, gastropod molluscs, and echinoderms - feed either as grazers, scavengers or carnivores. Grazing in this context relates to mode of feeding rather than diet, since there are few algae and grazers must subsist largely on an animal diet. Grazing is largely indiscriminate feeding as the organism moves over the rock surface, typified by sea urchins and some gastropod molluscs. Predation involves the deliberate selection, and sometimes pursuit, of prey. Starfish and decapod crustaceans are good examples. Usually predators seem poorly represented in hard-substratum communities, but they are often mobile and may only visit the biotope intermittently for feeding.

Circalittoral communities have been poorly studied in this respect. To determine the real role of species in a community generally requires experimental manipulation in the field. This has been done quite extensively in the intertidal, to a lesser extent in the infralittoral, but hardly at all in the circalittoral. It is largely a matter of the logistics of working at depth, but also the identification of significant questions to investigate. For hard substratum biotopes the most obvious questions usually relate to whether particular grazers or carnivores operate as 'keystone' species - i.e. their role is so significant that changes in their abundance can have major effects on community composition and functioning.

Keystone species are considered to be important in the maintenance of biodiversity by limiting the ability of potential dominant species to monopolise the available space. The effects of grazers and carnivores as keystone species will be discussed first, followed by a consideration of competition between species.

On intertidal rock limpets (Patella spp.) have such a role in Britain, and their removal can change barnacle-mussel dominated shores into fucoid algal dominated ones. This has been demonstrated experimentally (e.g. Jones, 1948; Hawkins, 1981), and was also seen following major limpet mortality after the Torrey Canyon disaster (Southward & Southward, 1978). In the infralittoral zone sea urchins play a similar role by grazing upon the algae and restricting the growth of kelps. In the Isle of Man it was found that removal of all of the common urchin Echinus esculentus from an area of rock substratum caused the lower boundary of the kelp forest to extend downwards by several metres (Jones & Kain, 1967). Off Norway sudden increases in the northern urchin Strongylocentrotus droebachiensis devastated the kelp beds to produce largely barren areas (Hagen, 1983) - examples of the classic 'urchin barren grounds'. It has been suggested that increased numbers of seals have reduced levels of the main urchin predator (the catfish Anarhichas lupus) allowing the urchins to proliferate (Sivertsen & Bjorge, 1980). In California reduction of lobsters which prey on sea urchins allowed urchins to proliferate and decimate the kelp canopies (Tegner & Levin, 1983). So there is evidence that grazing pressure on algae can moderate the depth of the infralittoral-circalittoral boundary, but the chain of events involved may be quite complex..

However, within the circalittoral zone itself there is less information on biological interactions, but sea urchins and starfish both have the potential to function in keystone roles. Sebens (1985a, b), working in the eastern USA, concluded that Echinus could prevent the development of the normal invertebrate community, maintaining a coralline crust which invertebrates would otherwise overgrow. This was confirmed by the experimental manipulation of urchin density, which also showed that certain species such as Alcyonium were less susceptible. Vertical and overhanging surfaces were grazed less effectively by urchins, and tended not to be reduced to the coralline crust status. Karlson (1978) and Vance (1979) have demonstrated a similar relationship in other American locations. In Norway, Sandness & Gulliksen (1980) excluded the urchin Strongylocentrotus with cages, and observed increased in barnacles and limpets. So urchins could perhaps also maintain a 'barren ground' scenario in the British circalittoral: in Scotland at least Echinus can reduce the diversity of the biota by intense grazing (Mitchell et al., 1983), and there is certainly scope for experimental study..

The starfish Asterias rubens is a common species and a major predator, particularly of mussels - it has even been used as a means of controlling mussel fouling on the legs of oil rigs (Ralph & Goodman, 1979). However, the effects of Asterias on common CFT species are less well known. It certainly preys heavily on the ascidian Ciona intestinalis, and can prevent this species attaining dominance (Gulliksen & Skjaeveland, 1973; Lundälv & Christie, 1986). It is also considered important in clearing space on rock by grazing barnacles, mussels and ascidians (Menge, 1982). On settlement panels in Sweden Asterias reduced the cover of sessile species to 20%, compared to 100% when they were excluded (Lundälv & Christie, 1986). Other predators include crabs, lobsters, and fish, but little is known of their roles in the British circalittoral. In S. Africa, Barkai & Branch (1988) excluded rock lobsters and observed marked changes. In Australia, Russ (1980) excluded fish, also with clear results.

Higher level predators are of interest because they may control the numbers of the major primary grazers and predators (urchins and starfish), so ultimately influencing community structure. In the eastern coasts of North America sea urchin populations have expanded over the past century, and this has been linked to the increase of lobster harvesting over the same period, on the assumption that lobsters were the most important urchin predators (Mann & Breen, 1972). However choice experiments indicate that urchins form a small part of lobster diet (Evans & Mann, 1977), and that crabs (Cancer borealis) form a major part (Sebens, 1985). Since crabs are important urchin predators, a decline in lobster numbers might in fact raise urchin predation. Relationships are complex, and currently unclear, but there is good reason to expect effects if the numbers of the top predators change (see section V.C).

There have been a great range of experimental studies, often using artificial substrata, investigating competition for space between sessile CFT species (see Sebens, 1985a, for a comprehensive list of references). Competitive success can result from a variety of mechanisms - physical and chemical aggression, bulldozing or smothering, shading, and perhaps localised food depletion. Erect species may limit competition for rock surface by having a small attachment area, resulting in niche partitioning (Jackson, 1977). A study of competion between rock wall species in Massachusetts over a two year period (Sebens, 1985a) enabled them to be ranked in a competitive hierarchy, with the colonial ascidian Aplidium and the plumose anemone Metridium at the top, and sponges and crustose algae at the bottom. Those species at the top of the hierarchy had thick and massive growth forms. These relationships were predominantly hierarchical, in contrast to the non-transitive ‘network’ relationships noted for cryptic coral reef fauna (Buss & Jackson, 1979). The implication is that undisturbed rock wall communities will progress to a ‘climax’ situation dominated by a few species, and that physical or biotic disturbance is important to generate and maintain diversity. In circalittoral biotopes physical disturbance seems to be uncommon, and biotic disturbance will be the major agent. Changes in abundance of keystone species could consequently have substantial effects on biodiversity. In Sweden areas from which starfish were excluded developed 100% cover of Ciona, but when they were present more species could co-exist (Lundälv & Christie, 1986).

Currently we do not know enough about biological interactions in British CFT biotopes. It is fairly certain that the space-occupying species are in a similar state of competition to that described by Sebens above - certainly competition was strong between species growing on artificial substrata (Hextall, 1994). The role of starfish, urchins and other higher level predators must be of similar importance to that demonstrated elsewhere. Relevant experimental studies are certainly needed, but as a precautionary measure major changes in abundance of potential keystone species must be taken seriously.

Next Section                      References