Читать книгу Marine Mussels - Elizabeth Gosling - Страница 51

Gastropods are significant predators of mussels worldwide. The dogwhelk, Nucella lapillus, is widely distributed on exposed shores in northern Europe and on the east coast of North America, where it feeds extensively on barnacles and small mussels. Predation is often seasonal, with whelks remaining aggregated in pools and crevices over the wintertime. However, numbers on mussel beds on the low and mid shore start to increase in the spring, and densities as high as 300 whelks m⁻² have been recorded over the summer months in NE England (Seed 1969). Profitability (energy assimilated from a food item relative to handling time) for dogwhelks feeding on mussels increases with prey size (Hughes & Dunkin 1984). Yet whelks prefer mussels smaller than the largest available. Hughes (1986) suggests that dogwhelks choose mussels with the maximum average profitability in the face of competition from other dogwhelks, which are attracted to the predator by olfactory stimuli from the damaged prey. Marples et al. (2018) recently suggested that profitability is a function of the current state of both the predator and the prey individuals, and that it should be considered to be an attribute of a particular encounter, in contrast to its present usage as an attribute of a prey species. The whelk uses the radula, a modified toothed chitinous structure, to drill a small hole through the thinnest part of the shell around the umbone or adductor muscle insertion regions (Seed 1976), or through the shell area overlying the glycogen‐rich digestive gland (Hughes & Dunkin 1984). Prior to drilling, the whelk softens the area using a secretion from the foot. The proboscis is inserted through the hole and the flesh of the prey is rasped away by the radula, converted into a ‘soup’ and devoured. Alternatively, a more efficient mechanism that is used by experienced whelks involves insertion of the proboscis through the valve gape and induction of muscular paralysis by injection of toxins. Feeding only occurs when conditions are conducive to such an activity, and during these times the dogwhelk consumes large quantities of food, so that the gut is always kept as full as possible. This allows shelter until more food is required, when foraging resumes. If waves are large or there is an excessive risk of water loss, the dogwhelk will remain inactive in sheltered locations for long periods. Feeding rates (drilling plus ingestion times) peak during the summer, but as water temperatures fall through the autumn, the time needed for ingestion lengthens, more than tripling the total handling time (Miller 2013). Prey handling time per mussel is generally in the range of two to three days, which agrees well with results from laboratory experiments showing that an adult whelk can consume about two mussels (1–3 cm shell length) per week during the summer (Seed 1969). Although this level of consumption may appear small, the high density of foraging whelks makes a serious impact on mussel coverage on exposed shores. For example, on rocky intertidal sites in Alaska, United States, where Nucella lima occurs at densities of >100 m⁻², Carrol & Highsmith (1996) estimated that the whelk can eliminate 60–90% of mussels (M. trossulus) at a given site in one season. Preference for mussels, as opposed to barnacles, appears to be fixed in early life: adult whelks transferred from sites with no mussel cover to those with a high coverage of mussels largely ignored the mussels, preferring to feast on barnacles (Wieters & Navarette 1998). Mussels ‘fight’ back by ensnaring and immobilising whelks in their byssus threads. An ensnared whelk can be overturned, thus arresting the drilling process (Figure 3.9). Whatever the fate of the mussel, the whelk, once ensnared, is trapped and exposed to predation (Petraitis 1987; Davenport et al. 1998; Farrell & Crowe 2007; Chiu et al. 2011). Sherker et al. (2017) were the first to use organisms raised in the field (Atlantic Canada), rather than in the lab, to demonstrate that predator‐induced morphological responses in bivalve prey hinder predation. During the spring and summer of 2016, they ran a field experiment that manipulated dogwhelk presence to test their nonconsumptive effects on mussel traits. Dogwhelk cues elicited thickening at the lip, centre and base of mussel shells while simultaneously limiting shell length growth. As shell mass was unaffected by dogwhelk presence, a trade‐off between shell thickening and elongation was revealed. Thickening was most pronounced at the thinnest parts of the shell. Using the field‐raised organisms, a lab experiment showed that dogwhelks took, on average, 55% longer to drill and consume mussels previously exposed to dogwhelk cues than mussels grown without such a cue exposure. Dogwhelks drilled at the thinnest parts of the shell, but nevertheless the consumed cue‐exposed mussels had thicker shells at the borehole than the consumed mussels not previously exposed to cues, which likely explains the observed difference in handling time. As handling time normally decreases predation success, this study indicates that the plastic structural modifications in mussels triggered by dogwhelk cues in the field hinder predation by these drilling predators.

Figure 3.9 Photograph of a whelk, Nucella lapillus, flipped on its back by a mussel, Mytilus edulis. Mussels attach a number of byssus threads to the body whorl of the predator gastropod, then retract them, flipping the predator over and immobilising it, and thus exposing it to crab predation.

Source: Photo courtesy of P. Petraitis, Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania, USA. (See colour plate section for colour representation of this figure).

On the West Coast of the United States, Nucella canaliculata and N. emarginata are major predators of mussels but seem to favour the thinner‐shelled species, M. galloprovincialis, to the thicker‐shelled and less nutritious M. californianus (Suchanek 1981). Other predatory gastropods such as Ocenebra poulsoni, Acanthina sopirata, Ceratostoma nuttalli and Jaton festivus also feed on Mytilus (Shaw et al. 1988), while various birds, crustaceans and fish feed on N. lapillus (details in Crothers 1985). Chemical cues are commonly used by prey to evaluate risk. Large et al. (2010) conducted a study to investigate the nature of cues used by prey hunted by generalist predators. Using Nucella and a suite of its potential predators as a model system, they explored how (1) predator type, (2) predator diet and (3) injured conspecifics and heterospecifics influence Nucella behaviour. Taking laboratory flumes, they found that N. lapillus responded only to the invasive green crab, Carcinus maenas – the predator it most frequently encounters – and not to rock crabs (Cancer irroratus) or Jonah crabs (Cancer borealis), which are sympatric predators but do not frequently encounter N. lapillus because they are primarily subtidal. Predator diet did not affect whelk responses to risk, although starved predator response was not significantly different from controls. Since green crabs are generalist predators, diet cues do not reflect predation risk, and thus altering behaviour as a function of predator diet would not likely benefit N. lapillus. However, the whelk did react to injured conspecifics – a strategy that may allow it to recognise threats when predators are difficult to detect. N. lapillus did not react to injured heterospecifics including M. edulis and herbivorous snails, Littorina littorea, suggesting that they respond to chemical cues unique to their species, allowing them to minimise costs associated with predator avoidance. The ability of prey to detect and respond to predator signals varies with environmental conditions (Large et al. 2011 and references therein).

Sea stars are also important predators that influence the distribution and abundance of mussels on the lower shore and in the sublittoral zone. Sea stars predate on mussels and other bivalves either by using force or by secreting an anaesthetic from their stomach that numbs the mussel and causes it to gape. They then extrude their stomach through their mouth into the shell opening and digest the prey. Sea stars with short or inflexible arms (e.g. Astropecten or Luidia spp.) usually swallow their prey whole while digestion occurs in the stomach, before egesting shell pieces through the mouth. Semmens et al. (2013) have reported a novel neurophysin‐associated neuropeptide that triggers stomach eversion and retraction, which may provide a basis for development of nonpeptidic small molecule agonists or antagonists that mimic or block the effects of neuropeptides; this could be used for chemical control of starfish feeding.

There are numerous studies of the sea star–bivalve interaction on rocky shores, primarily because of the ease of observation and manipulation of this pairing. Such studies have led to a greater understanding of the causes of zonation, and have provided additional evidence on size and spatial refuges in bivalve populations (Dame 2012 and references therein). On the Pacific coast of the United States, the mussel M. californianus exists as a well‐defined band on rocky intertidal shores. The sea star, P. ochraceus, is a major predator of this mussel (Figure 3.10). In a series of classic studies, removal of sea stars over a 10‐year period produced marked changes in zonation patterns, with a notable downward shift in the lower limit of mussel distribution of about 2 m through redistribution of adults and normal settlement of mussel larvae (Paine 1974). Consequently, the starfish has been regarded as a ‘keystone’ predator, one which through its feeding activities exerts a disproportionate influence on community structure, in this case setting the lower limits of mussel distribution. In a keystone predator‐dominated system, other invertebrate predators have minor effects on community structure, but in the absence of the keystone predator, such species may adopt a major role in the altered system (Navarrete & Menge 1996; see also Menge 1992; Menge et al. 1994; Naverette et al. 2000; Gosnell & Gaines 2012).

Figure 3.10 The sea star Pisaster ochraceus predating on mussels, Mytilus californianus, on the US Pacific coast.

Source: David Cowles, http://rosario.wallawalla.edu/inverts. Reproduced with permission. (See colour plate section for colour representation of this figure).

Until recently, the accepted explanation for the distinct zonation patterns on wave‐exposed rocky shores was that dense populations of sedentary organisms, such as mussels, form in static prey refuges above the reach of natural predators. Robles et al. (2009) showed that prey refuges are not in fact static. On the West Coast of the United States, experimental alteration of starfish (P. ochraceus) densities caused the downward extension of the lower boundaries of the mussel M. californianus, while experimental increases in sea star densities caused the upward recession of the lower boundary well into the zone presumed to be a spatial refuge for mussels from predation. As small mussel prey are depleted by sea stars over time, larger mussels are attacked, including those that would otherwise represent the lower boundary of their distribution. Based partly on what is known of Pisaster’s feeding behaviour, moving into the higher reaches of the mussel bed at high tide, then retreating to the lower parts of the bed at low tide, one might predict that the predators are more susceptible to stress than their prey. This was tested in Oregon by transfering mid‐tidal‐level mussels in April to high and low edges of the mussel bed and caging half of them with Pisaster (Petes et al. 2008). Three caging combinations were employed: mussels with and without sea stars, and sea stars without mussels. At four four‐weekly intervals in the summer, both mussels and sea stars were assessed for growth and reproductive status, and tissues were sampled for levels of Hsp70 (see Chapter 9). Results showed that mussels in the high‐level cages spawned earlier and had significantly higher levels of Hsp70 than those in the low‐level cages. There was no significant difference in growth of mussels at the two levels. Sea stars suffered higher mortality at the upper edge of the mussel bed, while those at the lower edge lost body mass but saw less mortality. Sea stars at either level exhibited no significant changes in the level of Hsp70. The authors concluded that intertidal stress factors, such as desiccation and temperature, affect the motile predator more than its sedentary prey. In a subsequent study, Hayne & Palmer (2013) found that P. ochraceus on wave‐exposed shores had narrower arms and were lighter per unit arm length than those from sheltered sites. Narrower arms probably reduce both lift and drag in breaking waves, while on protected shores fatter arms may provide more thermal inertia to resist overheating, or more body volume for gametes.

In northern Europe, Asterias rubens is a serious predator of M. edulis. This sea star aggregates seasonally on mussel beds in large numbers – sometimes as high as 450 m⁻² – often completely destroying local mussel populations (Dare 1982). In contrast to the results of laboratory‐based experiments, A. rubens shows no size selectivity when feeding in the field. The solid structure of interconnected mussels forming the bed, however, restricts predation to only those mussels situated at the bed surface, thus providing a refuge from predation for smaller mussels deeper down (Dolmer 1998). Not all mussels are equally susceptible to starfish predation. About 70% of M. edulis of North Sea origin were able to resist A. rubens, whereas all Baltic mussels (presumably M. trossulus) were opened within one hour (Norberg & Tedengren 1995; see also Lowen et al. 2013). M. edulis cultured in close vicinity to A. rubens were significantly smaller in shell length, height and width but had significantly larger posterior adductor muscles, thicker shells and more meat/shell volume. These morphological changes have an adaptive value in that predator‐exposed mussels have a significantly higher survival rate than unexposed mussels (Reimer & Tedengren 1996). Behavioural changes were also evident: predator‐exposed mussels in the laboratory formed larger aggregates, migrated less and sought structural refuges more often (Reimer & Tedengren 1997). Also, chemical aspects of epibionts, such as barnacles, algae, sponges and hydrozoans, on mussel (M. edulis) shells have the potential to modify the top‐down control by starfish (A. rubens) by changing or masking prey properties the sea stars cues upon, or by producing their own repellants (Laudien & Wahl 2004). This may explain why, in an earlier study, A. rubens was reported to prefer clean subtidal mussels over barnacle‐overgrown intertidal ones (Saier 2001). The most popular optimal foraging theory states that predators should maximise prey profitability (i.e. select that prey item that contains the highest energy content per handling time). However, Hummel et al. (2011) hypothesised that sea stars (A. rubens) do not forage on mussels (M. edulis) according to the classical optimal foraging theory, but actively avoid damage that may be caused by, for example, capture of foraging on too‐strong mussel shells; hence, they will have a stronger preference for mussels that are smaller than the most profitable ones. They found that A. rubens as a predator indeed chooses much smaller blue mussels to forage on than the most profitable ones. Hence, their study does not support the optimal foraging theory. There may be other constraints involved in foraging than just optimising energy intake. For example, predators may also be concerned with preventing potential loss or damage of their foraging body parts.

Sea star wasting disease or starfish wasting syndrome is a disease of starfish and several other echinoderms that appears sporadically, causing mass mortality of those affected. There are around 40 different species of sea stars that have been affected by this disease. It seems to be associated with raised water temperatures in some places but not others. It starts with the emergence of lesions, followed by body fragmentation and death. In 2014, it was suggested that the disease is associated with a densovirus now known as the sea star‐associated densovirus (SSaDV), although it is still not fully understood; see details in MARINe (2019).

Crabs (Cancer, Carcinus and Pachygrapsus spp.) are also significant predators of mussels on the lower shore and in the sublittoral zone. Their effect on mussel abundance is seasonal, with reduced predation in winter when crabs migrate offshore. Results from laboratory and field experiments show that crabs employ size‐selection of prey, with the upper size limit that can be opened being directly related to the size of the crab (references in Seed 1976). Crabs will almost always choose small‐sized prey when offered a range of sizes. It is handling time rather than the energetic costs of handling, estimated as a mere 2% of corresponding gains, that is the basis on which prey are selected (Rovero et al. 2000). During the handling period, the crab is at risk from other predators, competitors and even claw damage. Small mussels are therefore particularly vulnerable to predation, since they are easily crushed by most size classes of crab. A mussel must attain a shell length of at least 45 mm before it is relatively safe from crab predation. Once again, mussels show several defence mechanisms. In laboratory experiments, M. edulis increased byssus volume in response to waterborne cues from Cancer pagurus and Carcinus maenas (Cote 1995; Leonard et al. 1999). Similar findings were reported when P. viridis were exposed to crab (Thalamita danae) that had recently consumed conspecifics (Chiu et al. 2011). In addition, mussels subject to heavy predation develop thicker and more robust shells in response not just to crabs but also to the broken shells of other mussels (Leonard et al. 1999; see also Freeman & Byers 2006 and comments from Rawson et al. 2007 and Freeman & Byers 2007). Similar effects have also been reported in mussels subject to heavy whelk predation (Smith & Jennings 2000), and increased production of byssal threads has been shown in P. viridis in response to crab and gastropod predation (Cheung et al. 2004a,b, 2006, 2009). A behavioural strategy in response to crab predation has been reported in the Wadden Sea, Germany, where oysters (Crassostrea gigas) have invaded native mussel (M. edulis) beds (Eschweiler & Christensen 2011). Mussels subjected to direct contact with crabs (C. maenas) migrated from the top of the oyster reef to interspaces at the bottom of the reef, where they showed significantly reduced growth rates and conditions to mussels on the top of the reef. Mussels experience a trade‐off between survival and food supply, preferring to take refuge from predation even when this decreases growth and condition. In a previous study, Shin et al. (2008) investigated the effect of cues from damaged conspecifics and heterospecifics on the induction of refuge seeking and enhancement of byssus production as responses to the crab Carcinus maenas in the mussel Brachidontes variabilis, hypothesising that the mussel would seek refuge more readily and prefer a smaller refuge. More byssal threads (thicker and longer ones) should also be produced. They found that B. variabilis was able to differentiate between the sizes of available refuges and to stay in appropriate ones according to the level of risk they perceived. Staying in a smaller refuge would reduce the chance of the mussels being dislodged and consumed by the crabs. This helps explain why B. variabilis tended to stay in smaller refuges when predation risk was high, as simulated by the presence of damaged conspecifics and heterospecifics. In contrast, higher food and oxygen availabilities were found in large refuges, although the predation risk was also higher. Staying in larger refuges would be advantageous to the mussels only when predation risk was low. The preference toward certain sizes of refuge, therefore, should be a trade‐off between physiological requirements and the risk of predation.

Aquatic prey encounter an array of threat cues from multiple predators and killed conspecifics, yet the vast majority of induced defences are investigated using cues from single predator species. In most cases, it is unclear if odours from multiple predators will disrupt defences observed in single‐predator induction experiments (Freeman et al. 2009). Freeman et al. (2009) compared the inducible defences of M. edulis to waterborne odours from pairwise combinations of three predators representing two attack strategies, the sea star A. rubens, which pulls mussel shells open, and the crabs Carcinus maenas and Cancer irroratus, which crush or peel⁷ shells. Mussels increased adductor muscle mass in response to cues from unfed Asterias and increased shell thickness in response to unfed Carcinus. However, they did not express either predator‐specific response when exposed to the combined cues of Asterias and Carcinus, and did not increase shell thickness when exposed to cues from Cancer alone or any pairwise combination of the three predators. Shell closure or ‘clamming up’ did not occur in response to any predator combination. These results suggest that predator‐specific responses to Asterias and Carcinus are poorly integrated and cannot be expressed simultaneously.

Biotic invasions can result in the displacement of native species. This can alter the availability of native prey and the choices made by native predators. Skein et al. (2018) investigated prey selection by two native South African predators, the lobster Jasus lalandii and the starfish Marthasterias africana, in response to the invasive mussels M. galloprovincialis and Semimytilus algosus and the native mussels Aulacomya atra and Choromytilus meridionalis. As the diets of the two predators are broad, the authors hypothesised that they would consume the most abundant prey, regardless of its native or alien status. Laboratory studies presented predators with varying proportions of native and invasive mussels that represented pre‐ and post‐invasion scenarios. Both predators exhibited preference toward the native mussel C. meridionalis, even when it was the least abundant prey. The selection of native species occurred despite mussel parameters (shell strength, adductor muscle size and energy content), suggesting that invasive species would be easier to consume. These findings highlight the potential for facilitation of prey invasions, especially when predators avoid alien prey and select for native comparators that may offer resistance to the invasion through interspecific competition. This study does not present the first observation of native predators failing to select for invasive prey (e.g. López et al. 2010; Veiga et al. 2011). However, in other cases the avoided invasive prey were suggested to possess physical characteristics that might hinder predation from native predators (Skein et al. 2018).

Several bird species are predators of mussels. The main ones in western Europe that feed on mussels at low tide are the oystercatcher, Haematopus ostralegus, the common eider duck, Somateria mollissima, and the herring gull, Larus argentatus, although the latter only feeds on small mussels on newly established beds. Of these, the most significant is the oystercatcher (Figure 3.11). Predation is seasonal, with birds switching in the spring from mussels (and cockles) to deep‐living clams such as Scrobicularia plana and Macoma balthica, and back to surface bivalves in autumn in order to maximise intake rate (Zwarts et al. 1996a). The birds cannot survive if their diet is restricted to one or two prey species; they need to switch between three or four, and have to roam over feeding areas measuring at least some tens of km². Oystercatchers use one of two general techniques when preying on M. edulis: they either ‘stab’ their bills between the gaping valves of the mussel or ‘hammer’ through the shell on the dorsal or ventral side (Zieritz et al. 2012 and references therein). Stabbers and dorsal hammerers open the mussel in situ, but ventral hammerers usually tear the mussel off from the substrate and carry it to a suitably firm patch, where the shell is broken at the ventral region. They invariably select thin‐shelled mussels to hammer through because these are easier to crack than thick‐shelled mussels; it is the thickness of the prismatic layer that largely determines the vulnerability of mussel shells (Le Rossignol et al. 2011). Wintering oystercatchers feed extensively on M. edulis in the estuaries of southern Britain. They show a marked preference for brown‐shelled mussels over the commoner black‐shelled morph, and show that this enables them to maximise their rate of energy gain over a longer period than a single foraging bout (Nagarajan et al. 2002a). The brown and black mussels do not differ in ventral thickness or energy content, which are the main criteria for mussel selection and the most important for short‐term optimisation. The brown mussels contain significantly less moisture, so by selecting them, oystercatchers can pack more mussel flesh into their limited oesophageal storage capacity. This enables them to increase their overall consumption during a feeding bout and increases their long‐run energy gain rate, to an extent that is large enough to be significant for survival, especially during the short exposure of the mussel beds in winter. All birds show size selection within the prey species; this is because flesh content increases more steeply with prey size than handling time (Zwarts et al. 1996b). Oystercatchers are highly selective toward mussels of between 35 and 50 mm shell length, and fewer than 5% of mussels taken are below 35 mm or above 50 mm (Nagarajan et al. 2002b). The oystercatchers select ventrally thin‐shelled mussels, especially if the length is more than 35 mm. Removal of the largest mussels may reduce protected refuge for younger mussels, but may also allow younger mussels to grow at a faster rate – although, as already mentioned, gulls preferentially prey on small mussels (Goss‐Custard et al. 1996).

Figure 3.11 The American oystercatcher, Haematopus palliatus, a significant predator of bivalves, eating a surf clam (Spisula solidissima) at Nickerson Beach, Long Island, New York.

Source: ©Arthur Morris, www.birdsasart.com. (See colour plate section for colour representation of this figure).

Different diving duck species, including eiders (Somateria spp.), scoters (Melanitta spp.) and scaups (Aythya spp.), also predate extensively on mussels. In the case of eiders, mussels often constitute as much as 60% of their diet (Nehls & Ruth 1994). Eiders select mussels of smaller than optimal size, because this minimises shell ingestion, even though larger available prey would provide greater net energy gain per prey item (Bustnes & Erikstadk 1990; Hamilton et al. 1999). In the process of zoning in on their prey, the ducks may remove whole mussel clumps, thus causing mussel mortality over and above that produced by direct predation (Raffaelli et al. 1990).

Numerous studies have been undertaken to provide actual data on the impact of oystercatchers and other bird predators on commercial mussel beds. Analysis of faeces or regurgitated pellets provides information on the sizes of mussels selected by bird species. In the laboratory, the ash‐free dry weights (AFDW) of different sizes of mussels are determined in order to calculate the biomass eliminated (Hilgerloh 1999). According to exclosure experiments on the Danish tidal flats, oystercatchers and eiders eliminated 116 g AFDW m⁻² (Egerrup & Høegh Laursen 1992), while on the Dutch tidal flats, oystercatchers alone eliminated 48 g AFDW m⁻². Eiders consumed 360 g AFDW m⁻² on the west cost of Denmark. In a regional study in the Danish Wadden Sea, 300 g AFDW m⁻² emerged as a potential consumption, assuming that 100% of the food of all three predator species (eiders, oystercatchers and herring gulls) consisted of mussels and that eiders did not feed on mussels that lived submerged or on hard substrates (Faldborg et al. 1994). In the Wadden Sea of Lower Saxony, the same three bird predators consumed a total of 32 and 71 g AFDW m⁻² in two different years (Hilgerloh 1997). On a newly settled mussel bed in the same area, oystercatchers and herring gulls consumed 71 g AFDW m⁻² in five to six months (Hilgerloh et al. 1997). Predation pressure differs according to the available size classes and the predatory bird species, as birds are size selective.

It is well established that sea ducks feed in mussel aquaculture sites, but whether they are able to identify those mussels as being of higher quality or are only attracted by farms because of better accessibility of mussels (i.e. higher densities and convenient suspension in the water column) is not known. Varennes et al. (2015) showed that when detectability is controlled, eiders still choose the cultivated mussels. Preferences for cultivated mussels and their foraging advantages have important implications for sea ducks and habitat management.

Other birds that feed on mussels include knots, Calidris spp. (Alerstam et al 1992), and crows, Corvus spp. (Berrow et al. 1992a). Indeed, crows are significant predators of mussels in the intertidal zone and show several interesting adaptations. They frequently cache mussels during low tide, and recover them during high tide some two to three days later. This behaviour is believed to be a response to short‐term, daily fluctuations in food availability (Berrow et al. 1992a). In order to break them open, the crows drop mussels and other hard‐shelled prey on to hard surfaces such as roads or rocky shores (Berrow et al. 1992b). This behaviour peaks during October–February and usually involves only large‐sized mussels, no doubt an adaptation to food shortages in winter.

Mussel farms, with their very high densities of small, thin‐shelled mussels, can be foraging hot spots for diving ducks, particularly during spring and autumn when birds are building up their energy reserves for reproduction, migration or overwintering. For example, in spring 2011 in Baie de Chaleurs, Quebec, Canada, all mussel growers were severely hit by scoter predation, losing almost all their collectors and over 30% of their one‐ to two‐year‐old mussels on ropes (Varennes et al. 2013). Acoustic and visual deterrents have been tried with little success (Dionne et al. 2006). Provincial aquaculture authorities in Prince Edward Island, Canada proposed the use of protective socking material as a potential solution to the problem of diving ducks. The material consists of the standard polypropylene sock with a biodegradable protective layer stitched around it (Figure 3.12). When mussels are put into socks and hung in the water, they start migrating toward the outside of the sock in order to filter feed properly, making them more vulnerable to predation by diving ducks. The purpose of the second layer, with its smaller mesh openings, is to prevent their migrating outside the sock, keeping them between the layers until the bays freeze up in winter and the predation threat is over. Currently, in Canada, the United States and Europe, the only effective method that provides a complete and long‐term control of bird predation in culture facilities is the use of exclusion nets deployed around longlines and rafts of suspended mussel ropes (Varennes et al. 2013).

On the west coast of North America, the sea otter (Enhydra lutris) is an important predator of M. californianus. This species removes large clumps of mussels, which it sorts and consumes on the sea surface by pounding them on a flat stone on its chest or against other mussels. So, although sea otters are selective in terms of the size of prey they consume, they have profound effects on all size classes of mussel (Seed & Suchanek 1992). In addition, otters have substantial, indirect effects on the biomass of mussel bed‐associated communities. The total biomass of species associated with mussel beds was found to be more than three times higher where otters were absent (Singh et al. 2013). Other predators of mussels include sea urchins (Strongylocentrotus droebachiensis), lobsters (Panulirus interruptus and Homarus americanus), flatfish (Platichthys flesus, Pleuronectes platessa and Limanda limanda) and seals, walruses and turtles (see Seed & Suchanek 1992 for references).

Figure 3.12 Socking used for protection of mussels from predatory birds. (a) Regular socking materials used are the GDI‐4S, GDI‐5M and GDI‐7L (Go‐Deep International Inc., Fredericton, New Brunswick, Canada), for small, medium and large mussel seeds, respectively. These socks are made of interwoven, flattened polypropylene strands. (b) Protective socking material composed of regular (GDI) mussel socks with a biodegradable loop‐knitted sleeve in a 50:50 cotton:polyester blend sewn over the polypropylene strands, giving it its protective layer.

Source: From Dionne et al. (2006). Reproduced with permission from Springer Nature.

The most common pests of bottom‐dwelling mussels are shell‐burrowing sponges (Cliona spp.), polychaetes (Polydora spp.) and pea crabs (Pinnotheres spp.). The detrimental effects of pea crabs and boring polychaetes are described in Chapter 11.

Several management measures that prevent predation in mussel culture are described in Kammermans & Capelle (2019). In bouchot culture (Dardignac‐Corbeil 1975), crabs (Carcinus meanas, Maja brachydactyla) that predate on mussels can be prevented by placing a sheet around the bouchots. Predation by birds (e.g. gulls or diving ducks) on mussels on bouchots can be reduced by using nylon threads to prevent their landing. When sea stars and mollusc‐drilling snails (Nucella lapillus) are present in high densities and predation levels are high, they need to be manually removed.

Predation may exert a top‐down limitation on production, especially in bottom culture, since mussel plots are accessible for benthic predators as well as for fish and birds. Intertidal mussels are preyed upon by shore crabs and birds (oystercatchers, herring gulls), while subtidal mussels are preyed upon by shore crabs, sea stars and diving ducks. The number of sea stars on culture plots is reduced by freshwater treatment and there is a selective fishery on sea stars with sea star mops in The Netherlands, the United Kingdom, Germany, Ireland and purse‐seines (large walls of netting deployed around an entire area or school of fish) in Denmark (Petersen et al. 2016). Freshwater treatment is applied before seeding when mussels are in the vessels’ hold; the process consists of the joint exposure of mussels and associated sea stars to freshwater for several hours. Mussels will keep their shells shut, while sea stars are unable to protect themselves against osmotic stress and will not survive. Sea star mops are made of fuzzy rope entwined around small chains that are towed over the mussel plots ensnaring the sea stars, thereby enabling removal. Calderwood et al. (2016) estimated the efficiency of sea star removal by mops in Belfast Lough in Northern Ireland and found a large variation in the catch efficiency (4–78%), while the mean sea star reduction when applying this method was 27% (±SE 3.2.)

Studying the effect of exclusion of shore crabs in newly formed intertidal mussel beds on a scale of 800 m², Davies et al. (1980) found a 400–500% increase in yield over a period of two years. Rope or net culture of mussels has the advantage over bottom culture in that benthic predators cannot reach the mussels directly. Predation by mobile predators on mussels in raft or longline culture is therefore limited to diving birds and fish. However, predators with pelagic larvae (e.g. sea stars) commonly settle in long‐line farms. Ducks such as eider ducks that primarily feed on mussels can cause extensive damage to longline mussel cultures (Dunthorn 1971; Žydelis et al. 2009). In Maine, United States, mussels are protected by nets placed around the mussel rafts (Newell & Richardson 2014). Mussel ropes and nets are very attractive for a range of fish species. Due to a dramatic decline in Croatian shellfish production, Šegvić‐Bubić et al. (2011) investigated the most abundant fish species at a mussel farm situated along the eastern coast of the Adriatic Sea. Over a period of two years, the most abundant species observed were gilthead seabream (Sparus aurata), sand smelt (Atherina hepsetus) and mullet (Mugilidae species) in summer and autumn, A. hepsetus and bogue (Boops boops) in winter and the sand smelt (A. hepsetus), mullet (Mugilidae species) and saddled bream (Oblada melanura) in spring. At control locations, characterised by significantly lower fish assemblage abundances, the most abundant species were A. hepsetus and mullet (Mugilidae species) in summer, A. hepsetus and O. melanura in autumn, A. hepsetus and picarel (Spicara flexuosa) in winter and A. hepsetus and bogue (B. boops) in spring. Gilthead sea bream was extremely abundant at the mussel farm, with 5936 individuals censused in 155 of 192 fish counts (80%) and a maximum abundance of 285 individuals per 5000 m³. Stomach content analysis confirmed the presence of M. galloprovincialis as the dominant prey. During a single month, monitoring of 423 ropes revealed that approximately 828 kg of mussels with an average shell length of 34.3 ± 2.54 mm were destroyed within the first week of mussel deposition into the sea, highlighting the degree of fish predation on mussels at this farm. Šegvić‐Bubić et al. (2011) suggest that shifting farm concessions toward locations with greater depths may reduce fish predation.

Bivalves provide an excellent substrate for the settlement of many fouling organisms. Biofouling appears to be a significant cause of mortality in intertidal mussels, mainly due to dislodgement caused by the increased weight, especially from barnacles and seaweed. Fouling is a particular problem in suspended mussel culture, and almost 100 invertebrate species, including gastropods, crustaceans, bivalves, polychaetes, ascidians, sponges and hydroids, have been identified on mussel ropes (Hickman 1992). These organisms cause reduced growth and productivity through competition for space, but are not a major cause of mortality in suspended culture. See Chapter 10 for details on biofouling.

Mussels are the most prominent competitors for space in mid‐ to low‐shore areas on gently sloping rocky shores, but on steeper shores they tend to be replaced by barnacles or algae. Generally, where two mussel species coexist, there is competition but rarely elimination of one by the other. There are a few notable exceptions. M. galloprovincialis was accidentally introduced on the west coast of South Africa in the late 1970s, where it outcompetes the indigenous mussels Aulacomya atra and Choromytilus meridionalis by reason of its superior reproductive output, faster growth rate and greater tolerance to desiccation (Hockey & van Erkom Schurink 1992; Figure 3.13), while it exhibits partial habitat segregation with the local mussel, P. perna, on the south coast (Rius & McQuaid 2006). Because of weaker attachment strength in M. galloprovincialis, however, the species will be largely excluded from open coast sites, where wave action is generally stronger, although its greater capacity for exploitation competition through recolonisation will allow it to outcompete P. perna in more sheltered areas (especially in bays) that are periodically disturbed by storms (Erlandsson et al. 2006). Currently, South Africa is experiencing a second mussel invasion, with detection of the Chilean Semimytilus algosus in 2009 (de Greef et al. 2013). Both invasive species are now much more abundant intertidally than either of the indigenous mussels, Aulacomya atra or Choromytilus meridionalis, which have become largely confined to sublittoral and sand‐inundated habitats, respectively. The two invasive mussels display strong spatial segregation, with M. galloprovincialis dominating the midshore and S. algosus blanketing the lower shore. Alexander et al. (2015) predict that S. algosus will become established along the south coast of South Africa and that M. galloprovincialis will maintain dominance on the south and west coasts.

The mussels S. algosus and P. purpuratus cohabit most of the Chilean rocky shores, with the former inhabiting the low intertidal zone and the latter dominating the mid and mid to high zones. Field and laboratory experiments show that S. algosus is a weak competitor with respect to P. purpuratus, and post‐settlers present high mobility to relocate in the intertidal (Brante et al. 2019). Under this scenario, Brante et al. (2019) evaluated the dispersal behaviour of juveniles and adults of S. algosus as a potential response to competition with P. purpuratus. They also measured the attachment strength of S. algosus in the presence of its competitor as a measure of its escape response ability. Their results show that the presence of P. purpuratus increased the movement activity of juveniles and adults of S. algosus and decreased their attachment strength. Field experiments carried out with marked individuals on a Chilean rocky shore showed that S. algosus exhibits higher local dispersion in the zone where P. purpuratus is present. Mussels' high dispersal ability throughout the whole benthic phase may serve not only to reach the optimal physiological position in the intertidal, but also to reduce interspecific competition.

Figure 3.13 Performance of three indigenous South African mussels, Aulacomya atra, Perna perna and Choromytilus meridionalis, relative to Mytilus galloprovincialis. (a) Growth rate (mm in the first four years). (b) Total annual reproductive output as a percentage of body mass. (c) Survival rate after 24 months at a shore height experiencing 50% exposure to air per tidal cycle.

Source: From Branch & Steffani (2004). Reproduced with permission from Elsevier.

Another example of interspecific competition is the contribution of M. galloprovincialis to the displacement of M. trossulus along much of its historic range in southern California (Shinen & Morgan 2009). Other examples are those between M. californianus and the sea palm, Postelsia palmaeformis, on the Pacific coast (Dayton 1973; Blanchette 1996). The sea palm grows quickly and may overgrow M. californianus, eventually causing mussels to be torn free in the waves. This creates the space necessary for settlement of spores and recolonisation by Postelsia. Then there is the competition between M. galloprovincialis and the large indigenous limpet, Scutellastra argenvillei, on the west coast of South Africa, where the mussel is capable of forming dense, almost monospecific stands low on the shore. A survey indicated that at wave‐exposed locations, the abundance of M. galloprovincialis changes with exposure (Steffani & Branch 2003). At such locations, the mussel covered up to 90% of the primary substratum, whereas in semi‐exposed situations it was never abundant. As the cover of M. galloprovincialis increased, the abundance and size of S. argenvillei on rock declined, becoming confined to patches within a matrix of mussel beds. Both species were absent from sheltered shores and diminished where wave action was extreme. Comparisons with previous surveys indicated that exposed sites now largely covered by the alien mussel were once dominated by dense populations of the limpet. Therefore, the results of this survey provide circumstantial correlative evidence of a competitive interaction between M. galloprovincialis and S. argenvillei, and suggest that wave action mediates the strength of this interaction. The presence of mussel beds provides a novel settlement and living substratum for recruits and juveniles of S. argenvillei, albeit at much lower densities than in limpet patches. Adult limpets are virtually excluded from the mussel beds owing to their large size, which indicates the unsuitability of this habitat as a replacement substratum after competitive exclusion from primary rock space.

The American slipper limpet, Crepidula fornicate, was unintentionally introduced to Europe in the 1870s with oysters imported for farming purposes. Since the limpet is a filter feeder, trophic competition and associated negative effects when epizootic on bivalves have been assumed (Figure 3.14). Thielges et al. (2005) were the first to experimentally test the effects of C. fornicata on survival and growth of its major basibiont, M. edulis. In two field experiments (each lasting two weeks), epigrowth by C. fornicata resulted in a four‐ to eightfold reduction in survival of mussels, equivalent to a mortality of 28 and 30%, respectively. Shell growth in surviving mussels with attached C. fornicata was three to five times lower compared to unfouled mussels, but similar to that with artificial limpets. As a causative agent, interference competition in the form of changes in small‐scale hydrodynamics due to C. fornicata stacks was suggested. This could result in a high energy expenditure for byssus production of mussels. In general, interference and not exploitation competition is suggested to be the major impact of epizootic C. fornicata on its basibionts (organisms that are host to epibionts) in Europe.

Figure 3.14 Stack of four American slipper limpets, Crepidula fornicata, attached to the mussel Mytilus edulis.

Source: From Thieltges (2005). Reproduced with permission from Inter‐Research.

Typically, however, intraspecific competition for space is a more serious problem than interspecific competition, in that heavy spat fall of mussels on to adult beds can cause the underlying mussels to suffocate, thus loosening the entire population from the rock surface (Seed 1976). See Chapter 5 for information on interspecific and intraspecific gamete competition in Mytilus taxa.