Biodiversity & Conservation

Lophelia reefs



Image Murray Roberts - Section of Lophelia pertusa reef, Mingulay, Scotland. Image width ca XX cm.
Image copyright information

  • #
Distribution map

COR.COR.Lop recorded (dark blue bullet) and expected (light blue bullet) distribution in Britain and Ireland (see below)

  • EC_Habitats
  • UK_BAP

Ecological and functional relationships

Rogers (1999) stated that the ecology of Lophelia pertusa reefs was poorly understood and remained largely un-investigated. The following information is inferred from a few studies of the fauna of Lophelia reefs in nature and aquarium studies (Jensen & Frederiksen, 1992; Rogers, 1999; Mortensen, 2001). Although the major groups of organisms are probably similar, the exact species present will vary with location. Lophelia pertusa and other cold-water corals provide hard substrata for attachment of other epifaunal organisms, in the form of living and dead coral and coral fragments. The dense bushy growth of Lophelia locally modifies the environmental conditions (e.g. water flow) and provides a wide variety of niches for colonization by other species. Therefore, Lophelia may be regarded as an 'autogenic engineer' (Rogers, 1999) or key structural species.

Lophelia is a passive suspension feeder or passive carnivore, which has been observed to take zooplankton such as calanoid copepods and cumaceans in nature, and to take live zooplankton such as chaetognaths, small crustaceans (ca 1 mm e.g. copepods), and larger species such as krill ca 2 cm in length in aquaria (Mortensen, 2001; Mortensen et al., 2001). Mortensen (2001) demonstrated that Lophelia could also take a variety of foods, including dead food particles of krill, shrimp, herring and squid, and was able to reject unsuitable material including sediment. Mortensen (2001) concluded that Lophelia could utilize small organic particulates as food as well as live zooplankton. It probably also feeds on small invertebrates crawling over the coral surface.

The hard substratum provided by Lophelia, together with the strong currents in the areas it occupies, favours suspension feeding invertebrates, e.g. foraminiferans, sponges, hydroids, gorgonians (soft corals), corals, polychaetes, bryozoans, brachiopods, asteroids, ophiuroids, holothurians, and ascidians. For example, brittlestars, especially Ophiactis balli, were observed sheltering within the empty cups (calices) of dead corals with only their arms protruding (Rogers, 1999).

Mortensen (2001) suggested that there was a non-obligate mutualistic relationship between Lophelia and the polychaete Eunice norvegica, which shares a common distribution. In aquarium studies, Eunice norvegica was observed to steal food from the polyps of Lophelia, although Mortensen (2001) suggested that in nature Lophelia probably ingested live food before it could be stolen. Eunice norvegica was also observed to keep the coral surface clean of detritus and sedimentary particles, and in one instance attacked a sea urchin (Cidaris cidaris) that had climbed onto the coral. The polychaete may protect the coral from predators to some extent, a relationship seen in tropical coral communities (Mortensen, 2001). Most importantly, Eunice norvegica attaches its mucilaginous tube to the surface of the coral, which stimulates the coral to grow around and calcify the polychaetes' tube. This calcification may join adjoining coral branches, provide additional hard substrata for settlement of coral larvae and other invertebrates, and may strengthen the structure of the reef (Mortensen, 2001). In addition, tubes of the Eunice norvegica are capable of joining separate colonies, enhancing reef development, as seen in tropical corals (Dr Murray Roberts pers comm.).

The tubes of Eunice norvegica may also support other species of polychaete, e.g. the scale worm Harmothoe oculinarum.

Similarly, most bivalves were cavity dwellers occupying the empty calices of dead corals, e.g. Hiatella arctica and Acar nodulosa, while Delectopecten vitreus was found on the surface of live coral (Jensen & Frederiksen, 1992) and the giant file shell Acesta excavata may also use the coral as a substratum (Dr Jason Hall-Spencer pers comm.).

Jensen & Frederiksen (1992) observed only a few gastropods, the most numerous of which was Alvania jeffreysi, a predator of foraminifera. However, several species of gastropod have been recorded from Lophelia reefs (see Rogers, 1999), many of which are probably epifaunal grazers.

Many of the starfish and sea urchins recorded are probably epifaunal grazers and /or scavengers within the reef, while the mobile crustaceans including isopods, shrimp, crabs and hermit crabs are probably scavengers, or generalist predators of small invertebrates.

The coral skeleton may be eroded by several groups of organisms, e.g. bacteria, fungi, and sponges (e.g. Aka labyrinthica, Alectona millari and Cliona vastifera) which bore into dead corals, while eunicid, cirratulid, sabellid and spionid polychaetes also bore into the coral skeleton. Rogers (1999) noted that bioeroders play an important role in the development and maturation of coral reefs. Bioeroders reduce the coral skeleton to sediment, and weakens the coral structure so that pieces of coral break off or fall over. However, cavities produced by bioeroders also provide additional habitat complexity. Rogers (1999) noted that bioerosion in shallow coral reefs leads to rates of reef destruction that are only slightly slower than the rates of reef growth, so that any factor that reduces the growth rate of the corals may result in loss of the reef, especially since only a single cold-water coral species dominates this biotope.

Jensen & Frederiksen (1992) noted that many of the species they observed were only present as juveniles, suggesting that many species may use the Lophelia reef as a nursery area (Rogers, 1999).

Seasonal and longer term change

Lophelia reefs occupy relatively stable bodies of water (Rogers, 1999) but are still likely to experience seasonal fluctuations in current strength, temperature and food supply. The breaking of internal waves increases vertical mixing of the water column in areas of 'critical slope' (Frederiksen et al., 1997), which may occur close to the shelf break around the Faeroes Islands and the Faeroe-Shetland Channel interface (Roberts et al., 2003). Roberts & Anderson (2002a) noted that the polyps of Lophelia behaved asynchronously, without any clear diurnal patterns over a three day period in aquaria. Mortensen & Rapp (1998) detected distinct annual growth lines in Lophelia, and the growth of Lophelia from western Norway was carefully followed in aquaria over a 2.5 year period (Mortensen, 2001). Mortensen (2001) reported that linear extension of the skeleton was episodic, peaked in autumn, winter and spring, with a low growth period between June and September. In the aquaria, new polyps were generated mainly between August and December, the warmest part of the year, which suggested that temperature may be an important factor (Mortensen, 2001). However, deep-water population are probably not exposed to such temperature change. Mortensen (2001) observed no correlation between linear extension rates and temperature and salinity but concluded that the growth of the skeleton was correlated with seasonal variations in the abundance of particulate organic material and hence food availability.

The Lophelia reefs so far examined have been estimated to be extremely old, from several hundred to many thousands of years old. Therefore, although the longevity of individual coral polyps and associated species probably vary over time, the reef itself may be extremely long-lived (see 'time to reach maturity' below).

The associated fauna will probably exhibit seasonal fluctuations in abundance. For example, many bryozoan and hydroid species die back in the winter months. However, no other information was found.

Habitat structure and complexity

The shape and size of individual patches and reefs of Lophelia are highly variable, depending on local environmental conditions (Rogers, 1999). The density of branching varies and Lophelia may form robust 'bushes' in which the skeleton is thickened or finer more delicate branched colonies susceptible to damage e.g. from the pressure wave created by a submersible (Rogers, 1999). Reefs may be circular or 'halo-shaped', 'haystack-shaped', form domed mounds, or be elongated with one or more peaks, and the patches of reef may be arranged along the ridge of seamounts or banks in chains or 'islands groups' (Wilson, 1979a, b; Rogers, 1999; De Forges et al., 2000; Mortensen et al., 2001).

Wilson (1979b) suggested a model of Lophelia patch development, based on terms developed by Squires (1964), in which growth of an initial colony gives rise to coral fragments around it that either continue to grow or are colonized by Lophelia larvae. As the new colonies grow and merge they surround the central colony forming a 'thicket'. The central colony dies back, probably due to reduced water flow within the patch, and is reduced to coral debris, forming a halo shaped ring or 'coppice'. Subsequent phases of growth around the outside of the coppice results in concentric circles of growth forming a mature 'coppice' (see Wilson, 1979b for details). The reef becomes composed of several distinct zones, as exemplified by a Lophelia reef in the Stjernsund Fjord, Norway (Freiwald et al., 1997; Rogers, 1999). The living coral at the top of the reef grows on top of large fragments of dead coral, underneath which was a layer of small fragments and sediment. The living coral on top of the reef formed ring-shaped colonies as described by Wilson (1979b). Coral fragments from the main reef had also fallen down only to grow as spherical colonies. Away from the main reef Lophelia formed isolated coral thickets and dead collapsed frameworks (Rogers, 1999).

In the Darwin Mounds of the Rockall Trough, Masson et al. (2003) suggested that the mounds had preceded reef formation. In their study, Masson et al. (2003) observed no stratification of coral fragments in cores of the mounds, the cores being composed of quartz sand rather than bioclastic sediment. They concluded that mounds were formed by the deposition of sediment on the surface of the seabed by fluid escapes from the seafloor, and subsequently colonized by Lophelia and its associated fauna. The mounds form a raised substratum, which is a preferred habitat for Lopheliaand other suspension feeders (Masson et al., 2003).

The network of living and dead coral branches provide niches for a variety of organisms, e.g. bivalves and brittlestars within dead coral cups, and eunicids within and between the branches of corals (see above). However, the majority of the fauna observed were within and on the dead coral and coral debris (Rogers, 1999). Some coral mounds form acoustically detectable 'tails' aligned with the prevailing current, e.g. in the Darwin Mounds 'tails' included high densities of the giant protozoan Syringammina fragilissima (Masson et al., 2003).


Frederiksen et al. (1997) suggested that Lophelia reefs on the continental slopes off Norway, west Scotland and the Faroes, occupy a depth at which tidal currents impinge on raised seabed features with a critical degree of slope to generate internal waves. The resultant mixing of the water column above the shelf break generates nutrient rich surface waters, that in turn promotes phytoplankton productivity. Similarly, the increased mixing of bottom waters leads to resuspension of organic particulates from the seabed. Both effects can potentially increase the supply of food to Lophelia and other suspension feeders (Rogers, 1999). Rogers (1999) also noted that the massive reef complex on the Sula Ridge, off Norway was thought to rely on the supply of zooplankton from fertile surface waters. The occurrence of some Lophelia reefs in the vicinity of light hydrocarbon or methane seeps has led to the hypothesis that Lophelia reefs and their associated fauna may be supported by a chemosynthetic food chain (Hovland & Thomsen, 1997; Hovland, et al., 1998). But Rogers (1999) concluded that the evidence was equivocal. For example, occurrences of Lophelia in the Rockall Bank and elsewhere are not associated with hydrocarbon seeps (Rogers, 1999). Analysis of stable radiocarbon isotope (13C) levels in the skeleton of Lophelia pertusa and 13C/12C ratios in tissue is not consistent with a food chain based on hydrocarbon seeps (see Rogers, 1999 and Roberts et al., 2003 for discussion). Rogers (1999) suggested that most of the hydrocarbons are utilized by other organisms at the sediment-water interface.

Although, the only living part of the Lophelia framework are the surface colonies, the skeletal framework provides substratum, interstices, refugia and feeding grounds for a wide variety of other organisms. Most of the biomass of the reef is provided by the associated fauna, especially in smaller reefs (Rogers pers comm.). Overall, Lophelia reefs are probably highly productive ecosystems (secondary productivity) but no direct information was found. Lophelia reefs probably exhibit tight coupling between the pelagic and benthic ecosystems (Dr Murray Roberts, pers comm.).

Recruitment processes

Colonies of Lophelia grow by intratentacular budding, the division of an existing polyp into two polyps (Cairns, 1979; Rogers, 1999). In addition, Lophelia may generate new colonies by fragmentation, whereby coral fragments fall or are broken off, and continue to grow under suitable conditions. Fragmentation is a major mechanism whereby the initial colony expands to form a coppice and ultimately a reef (see 'habitat complexity' above and Wilson, 1979b). Some corals can reproduce by parthenogenesis, the development of an un-fertilized egg, while others exhibit 'polyp bailout' in which a polyp or piece of coral tissue leaves its skeleton, and moves to a suitable substratum and secretes a new skeleton (Richmond, 1997). However, there is currently no evidence for the existence of parthenogenesis or 'polyp bailout' in Lophelia (Rogers, 1999).

The mechanism of sexual reproduction in Lophelia pertusa is unknown (Rogers, 1999). About 25% of coral species have separate sexes while the majority are hermaphrodite, i.e. possess both male and female reproductive organs. In some species fertilization occurs externally and eggs and sperm are spawned into the sea. In other species fertilization occurs internally and the larvae develop and are brooded within the parent colony (Richmond, 1997). The larva of corals is the ciliated planula larvae. In species studied, the externally fertilized planula has been calculated to remain competent and capable of recruitment for 3-4 weeks. Brooded planula larvae tend to have a longer competence period, e.g. estimated to be over 100 days in Pocillopora damicornis (Richmond, 1997). The prolonged competency period is attributed to the provision of symbiotic zooxanthellae in brooded planulae that supplement larval energy reserves (Richmond, 1997). Cold-water corals, such as Lophelia, lack zooxanthellae, however Roberts (2002a) noted that larval competency in cold temperate waters may be considerably longer than observed in tropical waters.

The planula larvae of Lophelia pertusa require hard substrata for settlement, including rock surfaces, artificial substrata, coral fragments or hydrocarbon seep associated carbonates. In sedimentary areas, Lophelia pertusa may settle on hard substrata as small as a shell, pebble, or worm tube (Rogers, 1999). However, a hard substratum is a pre-requisite for settlement and a layer of sediment may interfere with settlement and hence recruitment.

Whilst Rogers (1999) noted that there was no indication of the dispersive capability of Lophelia pertusa, its ability to colonize isolated hard substrata and artificial substrata such as submarine cables, the Brent Spar storage buoy and oil rigs suggests that it has a pelagic larval phase (Rogers, 1999; Roberts, 2002a). Roberts (2002a) concluded that the occurrence of Lophelia on structures in the Beryl and Brent oil fields in the North Sea was good evidence for a dispersive planula larva. Roberts (2002a) suggested that the colonies in the North Sea oil fields originated as larvae from the offshore banks of the Atlantic margin, and were carried into the North Sea in cooled Atlantic water, possibly via the east Shetland Atlantic Inflow current. Transport of larvae in the water mass of prevailing water currents probably provides the opportunity for long distance dispersal.

The recent evidence (above) suggests that larvae are dispersive but that migration is not sufficient to counteract reproductive isolation of populations (Dr Alex Rogers pers comm.). Molecular genetic data is somewhat confused at present but microsatellite data indicates that Beryl oil fields samples of Lophelia are closely related to northern Rockall Trough populations and that there is strong genetic differentiation (population sub-division), with very low gene flow between areas (Dr Alex Rogers pers comm.). Present evidence suggests that asexual reproduction predominates in reef growth and that the contribution from larvae may be limited. Therefore, molecular genetic data suggests that recolonization of a disturbed areas is likely to be slow (Le Goff-Vitry & Rogers, 2002, summary only, Dr Alex Rogers pers comm.).

The associated epifauna and interstitial fauna probably depend on locality and recruit from the surrounding area. Many hydroids, ascidians and probably sponges have short lived planktonic or demersal larvae with relatively poor dispersal capabilities. Exceptions include Alcyonium digitatum and hydroids that produce medusoid life stages and probably exhibit relatively good dispersal potential. Hydroids are opportunistic, fast growing species, with relatively widespread distributions, which colonize rapidly and are often the first groups on species to occur on settlement panels. Sponges may take longer to recruit to the habitat but are good competitors for space. Recruitment in epifaunal communities is discussed in detail in the faunal turf biotopes MCR.Flu, CR.Bug and in Modiolus modiolus beds (MCR.ModT). Mobile epifaunal species, such as echinoderms (starfish and brittlestars), crustaceans, and fish are fairly vagile and capable of colonizing the community by migration from the surrounding areas. In addition, most echinoderms and crustaceans have long-lived planktonic larvae with potentially high dispersal potential, although, recruitment may be sporadic, especially in echinoderms.

Time for community to reach maturity

Mortensen et al. (2001) suggested that the size of Lophelia reefs was determined by the time taken for development and the topography of the seabed that affects both the area over which coral fragments and rubble can spread and the local currents and hence, food supply and growth rates.

The growth rate of Lophelia is very slow. Estimates of growth rate range from 2 to 25 mm/yr. depending on location (Wilson, 1979b; Rogers, 1999; Hall-Spencer et al., 2002; Roberts, 2002a) although inaccurate sampling of the coral skeleton may have led to biased estimates. Studies of growth lines suggested a mean extension rate of 5.5 mm/yr., with linear extension rates greatest in the early stages of polyp growth, slowing with age (Mortensen & Rapp, 1998). Measurement of linear extension rates in aquarium specimens gave a mean annual growth rate of 9.4 mm/yr. (Mortensen, 2001). Rates of growth on artificial structures were estimated to range from 6 mm/yr. on submarine cables in north west Spain to 26 mm/yr. on the Brent Spar storage buoy (Bell & Smith, 1999; Roberts, 2002a).

Estimates of potential age of Lophelia colonies and reefs vary with location and with the growth rates estimates used to calculate age. For example, Wilson (1979b) estimated that a single colony 1.5 m in height would probably be 200 -366 years old (based on a growth rates between 7.5 and 4.1 mm/yr. respectively). Lophelia reefs sampled off Norway, 25 m in height and 330 x 120 m in area were probably between 1,000 and 6,250 years old, depending on growth rate (Rogers, 1999). Radiocarbon dating of cold-water corals from west Ireland, provided estimated ages of 451 years before present (BP) for live Lophelia pertusa and 762 years BP for dead Lophelia pertusa fragments (Hall-Spencer et al., 2002). However, dead coral rubble formed by the cold-water coral Desmophyllum cristagalli, at the same site, were between 4067 and 5001 years BP, which suggested that the reef system was probably at least 4,500 years old (Hall-Spencer et al., 2002). The age of Lophelia reefs in south east Norway and west of Fedje Island, west Norway was estimated to be 8,700 and 3,600 years BP respectively (Mikkelsen et al., 1982; Rokoengen & Østma, 1985; Mortensen et al., 2001). The Lophelia reefs of the Sula Ridge were estimated to be 8,600 years old, having developed over the last 10,000 years since the last ice age (Hovland & Mortensen, 1999; Mortensen et al., 2001). Coral rubble from cold-water coral reefs on the Florida Hatteras slope, which was not ice covered, had an age of ca 20,230±230 years BP (Mortensen et al., 2001).

Recruitment to available hard substrata by epifauna such as hydroids, and ascidians is probably fairly rapid (see MCR.Flu or CR.Bug), with sponges and soft corals taking longer to develop. Bryozoans, hydroids, and ascidians are opportunistic, grow and colonize space rapidly and will probably develop a epifaunal cover within 1-2 years (for example see Sebens, 1985, 1986). Mobile epifauna and infauna will probably colonize rapidly from the surrounding area. Slow growing species such as some sponges and anemones (see Sebens, 1985, 1986), will probably take many years to develop significant cover, so that a diverse community may take up to 5 -10 years to develop, depending on local conditions.

While, epifaunal and infaunal species would colonize relatively rapidly, the key species determining the development of the reef is Lophelia itself. Deep-sea communities are thought to have very slow colonization rates (Rogers, 1999). While Lophelia may have a dispersive larval stage (see Roberts, 2002a), there is little information available on recruitment rates in natural systems. However, recent molecular genetic data suggests that larval recruitment is probably low or sporadic (Le Goff-Vitry & Rogers, 2002, summary only; Dr Alex Rogers, pers comm.). Overall, even with good recruitment, Lophelia is very slow growing and would probably take several hundred years to develop large colonies (ca 1.5-2 m in diameter) and several thousand years to develop a reef system 10 -30 m thick (Fosså et al., 2002).

Additional information

None entered.

This review can be cited as follows:

Tyler-Walters, H. 2005. Lophelia reefs. Marine Life Information Network: Biology and Sensitivity Key Information Sub-programme [on-line]. Plymouth: Marine Biological Association of the United Kingdom. [cited 18/04/2014]. Available from: <>