Wild seaweed harvesting: strategic environmental assessment - environmental report

6. Effects of Harvesting on Ecological Function

6.1. Introduction

6.1.1. The potential effects of wild seaweed and seagrass harvesting on the ecological function of seaweeds and seagrasses are reviewed and discussed in this section. Mitigation measures that need to be considered to ensure that wild harvesting activities are sustainable are presented in Section 9 .

6.1.2. Wild harvesting activities have implications for the population structure, community dynamics and wider functioning of marine ecosystems (Smale et al, 2013). They may also affect ecosystem services. Potential issues include but are not limited to:

• loss of habitat and/or shelter for a range of plants and animals, alongside loss of direct and indirect food sources. As well as detrital grazers, suspension feeders et al, this has consequences for higher trophic levels, e.g. mammals, birds and piscivorous fish.
• loss of nursery grounds for juvenile invertebrates and fish, with consequences for higher trophic levels and commercial fish stocks.

6.1.3. These effects depend on a range of factors, including but not limited to:

• the species to be harvested,
• the harvesting method,
• the amount taken,
• the harvesting location and its environmental context,
• the time allowed for regeneration prior to harvesting again, and.
• the timing (season) of harvest.

6.2. Effects of Harvesting

6.2.1. The resistance (tolerance) and resilience (recovery) of seaweed and seagrass species to harvesting and the subsequent effect this activity has on the wider ecological community has been subject to many studies in the UK and abroad. The following sections review the potential ecological effects of wild harvesting on each of the broad groups of seaweeds and seagrasses. The effects of harvesting on the target seaweed or seagrass resource itself ( i.e. the biotope) is initially reviewed followed by a review on other ecological functions ( i.e. ecological interactions, food web dynamics and production) where information is available.

6.2.2. Effects on seaweeds and seagrasses depend on the harvesting methods used and the habitat requirements of the particular species.

Wracks

6.2.3. Ascophyllum nodosum has a long life span; individual fronds can survive for 10 to 15 years and assemblages originating from a common holdfast are thought to be capable of living for decades (Holt et al., 1997 and references therein). The evidence on recovery rates from natural disturbance or harvesting does not agree.

6.2.4. Recovery of very sheltered shores dominated by Ascophyllum spp. from natural disturbance may take decades (Hill and White, 2008). Early studies on Ascophyllum showed a failure of this species to fully recolonise harvested or experimentally manipulated areas up to eleven years later (Knight & Parke, 1950; Boney, 1965; cited in Jenkins et al., 2004). Jenkins et al. (2004) investigated the effect of experimental Ascophyllum canopy removal over a twelve-year period. The Ascophyllum canopy was slow to recover, with no recovery after 6 years and 46% coverage after 12 years (compared to 80-100% cover in uncleared plots). Removal of the canopy led to short term changes in the community composition (namely reduced cover of red algal species and increased area grazed by limpets) which were still apparent 12 years later. After 12 years the affected areas were dominated by other wrack species, namely Fucus serratus and Fucus vesiculosus.

6.2.5. Harvesting of A. nodosum is commonly carried out in most areas of its distribution. For example, in the Western Isles it is reported that Ascophyllum is hand cut using a sickle and cutting occurs all year round (The Minch Project website). Mechanical harvesting using a seaweed harvesting boat also occurs in Scotland and is currently restricted to the Outer Hebrides ( Table 6).

6.2.6. Harvesting of Ascophyllum nodosum involving the removal of the entire plant would severely affect the population given the species' slow growth rate and poor recruitment ( e.g. Holt et al., 1997). However, a study in Ireland reported that if stumps 10-20 cm high are left, the plants can re-sprout and re-harvesting is possible after 3 to 6 years (Guiry, 1997; cited in McLaughlin et al., 2006).

6.2.7. Seaweed harvesting boats can only operate when there is a safe depth of water underneath and therefore mechanical hedge cutting is likely to remove less of the plant than hand cutting methods (Walter Speirs, Scottish Seaweed Industry Association, pers. comm.). However, the scale of removal by mechanical means is likely to be greater than by hand harvesting alone.

6.2.8. In an area of Strangford Lough where harvesting of Ascophyllum was carried out on a small scale, ecological effects were found up to 3 years after the harvesting ceased (Boaden & Dring, 1980; cited in Hill, 2008 and McLaughlin et al., 2006). In the cut area, the growth rate of A. nodosum increased but shore coverage was reduced. The cover of green algae and F. vesiculosus increased as did the density of grazers, namely limpets ( Patella sp.). Microalgal cover of boulders had also increased and had significantly more crustacean meiofauna. It was concluded that harvesting Ascophyllum even at a small scale has a significant and persistent effect on shore ecology (Boaden and Dring, 1980).

6.2.9. Another study was undertaken on the impact of hand and mechanical harvesting of A. nodosum at two sites on the West coast of Ireland. Hand harvesting involved cutting floating fronds by hand (from a boat) leaving about 30% cover and 20 cm length of each plant. In both the hand and mechanically harvested areas, most plants were harvested. Kelly et al. (2001) reported that there was no overall impact on the biodiversity of the harvested sites and the percentage cover of Ascophyllum was nearing recovery after 11 to 17 months. Although the proportion of F. vesiculosus cover increased at the harvested sites, there was no effect on other species of macroalgae, fish populations or other large epifaunal species. These findings correlate with studies in Nova Scotia which found that the impact of harvesting 95% of the A. nodosum standing stock resulted in no decrease in fish abundance or diversity (Tyler, 1994).

6.2.10. After harvesting, fucoid species such as Fucus spiralis and F. vesiculosus both rapidly recruit cleared areas (Holt et al., 1997; cited in White, 2008b; 2008c), with full recovery of F. vesiculosus taking 1 to 3 years (Holt et al., 1997; cited in White, 2008c). Fucus spp. can regenerate from the remaining stem provided that it is not removed entirely (White, 2008c). The spores of F. serratus are broadcast into the water column allowing a potentially large dispersal distance. Recruitment occurs through reproduction of the remaining population or from other populations. It was concluded by Jackson (2008) that if some of the Fucus population remains it is unlikely that other species will come to dominate; however, if the entire population is removed, other species may establish and dominate. Re-establishment may depend on the ability to out-compete other species and this in turn may be dependent on suitable environmental conditions. Recovery from disturbance (such as abrasion, physical disturbance and hydrocarbon contamination), where some of the population remains, is likely to occur after a year (Jackson, 2008).

6.2.11. Another study in Spain noted the benefits of shading effects on the survival of fucoids. An increased physiological resilience to low tide stressors (namely desiccation and irradiation) was found in covered F. serratus and F. spiralis plants compared to uncovered plants (Fernández et al., 2015). Therefore, the harvesting of fucoids or other seaweeds would remove these protective shading effects.

6.2.12. Recovery rates of Pelvetia canaliculata after harvesting may be variable (White, 2008a). Subrahmanyan (1960; cited in White, 2008a), for example, observed that this species readily recruits cleared areas of the shore with full recovery of the community taking place within 5 years. However, in Shetland P. canaliculata did not recolonise shores that had been bulldozed until 7 to 8 years after the event. Overall therefore, wracks are reported as having moderate recoverability following extraction (White, 2008a).

Kelps

6.2.13. Each species of kelp has a different growth rate, growth season and life-span (see Appendix C, Table C1). Different populations of the same species may also behave differently. Laminaria hyperborea plants can live up to 25 years in Iceland (Gunnarsson, 1991 cited in Wilkinson, 1995) and up to 10 to 15 years in Norway (Sjøtun et al., 1993 cited in Wilkinson, 1995; Steen et al., 2015). The lifespan of this species in Scottish waters is generally 5 to 7 years, with occasional 12 to 15 year old plants also having been reported (D. A. Macinnes, Marine Biopolymers, pers. comm.).

6.2.14. Most of the upper part of the frond or blade of a kelp plant can be removed and the blade will slowly re-grow, but if the growth area of the blade (the meristem at the junction of the stipe and frond) is damaged or removed, the stipe and holdfast degenerate and the whole plant dies as regrowth cannot occur (Birkett et al., 1998). In the case of hand harvesting where the plant is cut below the meristem and the stipe is left in place after cutting, toxic compounds are excreted as the stipe decays which in turn hinders spore germination and plantlet growth (Kelly, 2005). In addition to the decaying compounds, another problem associated with leaving the stipes is that a calcareous film forms on the surface of the substrata. Although this provides a surface for kelp plantlets to settle and grow, this surface is not robust enough to support larger plants which are then ripped away during a storm event or strong currents (Kelly, 2005).

Kelp regeneration

6.2.15. Recovery from damage and/or removal and the rate of kelp regeneration will depend on a number of factors including the life history characteristics of the species affected, the area of the plant cut, nutrient availability, irradiance levels, the level of wave exposure and the presence of grazers ( CDFG E.I.R., 2001; Kelly, 2005; Sjøtun et al., 2000). For example, studies in Norway have found that the growth rates of L. hyperborea are higher in wave exposed locations (Sjøtun et al., 2000). Therefore, harvested kelp can recover more rapidly in wave exposed locations than in sheltered locations.

6.2.16. In an assessment of benthic species' sensitivity to fishing disturbance, MacDonald et al. (1996) classified the kelp species L. hyperborea (mature) as having 'moderate' recovery potential. Using a similar methodology, McMath et al. (2000) scored the recruitment ability of kelps as 1-20 (on a scale of 1-100, where '1' represents the maximum recruitment success and 100 represents no recruitment ability) based on life history characteristics (rapid growth rates of 1 to 5 cm/week, sexual maturity at 1 to 2 years and frequent reproduction). The regenerative ability of kelps was 'scored' as 20 to 30 (out of a scale of 1-100, where '1' represents the maximum regeneration ability and 100 represents no regeneration ability) as rapid re-growth of kelp blades can occur following damage and/or removal, providing the meristem remains intact ( ABPmer, 2013).

6.2.17. Information on the recovery of Laminaria from disturbance and removal is provided both by experimental kelp removal studies and from observations at harvested grounds.

6.2.18. Experimental canopy removal and clearance experiments conducted in Scotland and the Isle of Man showed that 3 years after canopy removal, some semblance of a kelp forest, in terms of macroalgal biomass and subsidiary algal species, was regained (Birkett et al., 1998). However, the size of the kelp plants and age structure of the population was different from uncleared kelp forests. These experimental clearing experiments, however, do not directly mimic the effect of mechanical harvesting by which the kelp stipes would be removed.

6.2.19. On the Isle of Man, studies by Hawkins & Harkin (1985) and Smith (1985) observed the effects of the removal of L. digitata and L. hyperborea. It was found that L. digitata re-grew whereas L. hyperborea did not (Kelly, 2005). Kain (1975) examined the successional recolonisation of seaweed in areas of L. hyperborea forest that had been cleared. After 2.5 years L. hyperborea was again the dominant species with red algae also present, resembling a very similar seaweed composition to that present prior to the clearing (Kain, 1975).

6.2.20. Experimental work in Nova Scotia where Laminaria longicruris and L. digitata are harvested has shown that if kelp beds are destroyed and/or partially destroyed, grazing sea urchins may prevent regeneration and recruitment of climax kelp communities. It is thought that kelp harvesting removes the cover and protection of urchin predators ( e.g. lobsters, crabs and fish) and this consequent reduction in predator pressure enables increases in urchin populations which then graze destructively on Laminaria, resulting in areas devoid of kelps (Bernstein et al., 1981; reviewed in Birkett et al., 1998). In Scotland, urchins do not tend to eat adult kelps but hinder re-establishment on cleared areas by grazing sporelings (Wilkinson, 1995). In addition, following harvesting in areas where there is an established population of urchins, the urchins can function as detritus feeders and remain at a sufficient level to inhibit kelp regeneration (Warner, 1984 cited in Wilkinson, 1995).

6.2.21. A large number of studies have been undertaken in Norway to monitor the effect of kelp harvesting ( e.g. Birkett et al., 1998 and references therein; 2014a; 2014b; Steen et al., 2015). These indicate that harvested forests of L. hyperborea recovered kelp biomass within 2 to 4 years but that individual kelp plant sizes were still below pre-harvesting levels 4 years later. In addition, significant differences in the understorey density, epiphyte community, epifaunal species, holdfast fauna and benthic macrofauna and flora persisted 4 years after harvesting. The recovery of the kelp in terms of size of the kelp plant and number of epiphytes was more rapid in wave exposed areas.

6.3. The kelp forest habitat

6.3.1. A study by Christie et al. (1998) looking at the effects of kelp harvesting on epifaunal communities in Norway concluded that recolonisation by fauna depends upon the recovery time of the kelps and of the fauna. Overall, it took 6 years following the harvest for faunal abundances to stabilise (Christie et al., 1998).

6.3.2. A comparison of kelp harvesting methods reveals that certain types of trawling devices, that involve the entire removal of larger mature plants, result in a predominance of younger plants. Trawling is considered to open the area up to high levels of recruitment as a result of increased levels of light availability (Christie et al., 1998; Kelly, 2005). This recruitment ensures the persistence of a kelp forest and does not give space for other opportunistic species (Christie et al., 1998); little degradation of the habitat therefore occurs. In addition, younger plants have a higher percentage of alginic acid and are of a better end-use quality than older kelps. Although this is preferable when considering maximum yield, from an ecological point of view this will result in homogenous populations of younger plants growing on regularly disturbed substratum. If mechanical harvesting occurs over a number of years it is conceivable that only the dominant, fast growing species will be present (Kelly, 2005). This could result in a reduction in habitat complexity, biomass and diversity of an area which in turn reduces the seaweed capability as a habitat, shelter and food source for a number of species.

6.3.3. The ability of a kelp forest to provide a habitat, shelter and food source for a wide range of species will be hampered by the kelps being harvested. The removal of kelps would affect many marine fauna and flora that use this habitat as feeding and nursery grounds. Some short term studies have reported that the harvesting of kelps may negatively affect fish recruitment. The number of juvenile gadoids was significantly reduced or not present in newly harvested areas and continued to be reduced for at least 1 year following harvest (Steneck et al., 2002; Sjøtun & Lorentsen, 2000). Also, a recent study in Norway found a significant reduction of small cod and an increase in wrasse two years after kelp harvesting (Bodvin et al., 2014a). Conversely, a number of other studies in Norway found no significant effects of kelp harvesting on fish or crab catches (Steen, 2010; Steen et al., 2013; Bodvin et al., 2014b; Steen et al., 2014; Steen et al., 2015). Such effects may however be disguised by large variations in the data sets, for example seasonal variations in fish and crab abundance (Bodvin et al., 2014b). Furthermore, no long term studies have been carried out and therefore the period for full recovery is not known. Although no direct evidence of this impact has been reported, the harvesting of kelp habitat would also remove the availability of food to higher trophic levels including seabirds (Steneck et al., 2002).

6.3.4. It is important to ensure that harvesting of climax seaweeds does not tip the community to alternative, less ecologically valuable climax communities or to opportunist communities (see Section 3.3). Part of the mechanism by which climax communities can maintain themselves is for the dominant species to be able to replace dying older plants with younger ones of the same species. In Laminaria forests this can occur because the young plants may be kept at a stage of arrested development in the shade cast by the canopy of the older plants (Wilkinson, 1995). Removal of the canopy, which could be due to natural death or due to harvesting, enables the younger plants to grow to replace the older ones. Advantage can be taken of such processes in designing harvesting strategies in order to preserve the resource.

6.4. Kelp harvesting methods

6.4.1. Mechanical cutting removes all kelps, irrespective of size (Vea & Ask, 2011). The removal of juvenile plants may allow opportunistic species to move into the habitat. These species can inhibit the regrowth of kelps (Scheibling & Gagnon, 2006) and may cause a decrease in species richness (Wells et al., 2007) ultimately resulting in long term habitat degradation.

6.4.2. There are a number of other ways in which the harvesting methods can be modified to maximise recovery rates and recruitment. A system of rotation of harvested areas was introduced in Norway to ensure that each area of kelp forest was harvested only once in 4 years to allow the kelp to regrow. It has since been recommended that this timescale be extended to 7-10 years to allow for the partial recovery of populations of non-kelp species (Birkett et al., 1998).

6.4.3. Recruitment success is also found to be enhanced if harvesting of kelps is carried out in patchy patterns, allowing for recruitment from the surrounding, non-harvested areas (Norderhaug et al., 2003; Waage-Nielsen et al., 2003). L. hyperborea spores, for example, can disperse over 200 m and therefore recruitment success reduces in harvested areas that span more than this distance (Frederiksen et al., 1995). Algal spores remain viable in the laboratory for 4 to 11 days (Hoffmann & Camus, 1989), which is sufficient to allow them to drift a considerable distance. In the sea, if not eaten, propagules may last even longer, as they are able to photosynthesise (Kain, 1964; McLachlan & Bidwell, 1978; Amsler & Neushul, 1991, Norton, 1992). Within the laboratory undeveloped gametophytes of L. hyperborea, L. digitata, Saccharina latissima and Saccorhiza polyschides have all been recorded as being able to survive in the dark for at least 80 days (Kain & Jones, 1969).

Red Seaweeds

6.4.4. Red seaweed species such as Chondrus crispus and Mastocarpus stellatus regenerate at the holdfasts, from the surface and edges of the severed fronds, and also by recolonisation of sporelings (The Minch Project website). Depending on the method of harvesting, recovery for these plants can be as rapid as 6 months for raking and 18 months for a closer crop (The Minch Project website). When cut by hand and the holdfasts left intact, the red seaweed Porphyra spp. has been recorded as having complete biomass recovery within 60 days. However, if the holdfast is removed, biomass recovery is very limited (The Minch Project website). Anecdotal evidence indicates that Porphyra spp. and Palmaria palmata have been completely harvested from certain areas by hand (Juliet Brodie, Natural History Museum, pers. comm.)

6.4.5. In addition, the timing of the harvest can affect reproduction and recoverability of seaweeds. For example, it was found in New Hampshire if M. stellatus was harvested in August and the holdfasts were not damaged, plant biomass could be re-established by the following July (The Minch Project website).

6.4.6. Therefore it is concluded that, although recovery times will vary between species, on the whole recovery of red seaweeds is generally quick, taking place within 18 months. In order to maximise recovery rate, the holdfast needs to be left. The holdfast provides an area for spores to settle and recolonise which will also increase recovery rates. In addition, the holdfast of red seaweeds provides a habitat and shelter for supporting species and so disturbance to the wider community would be minimised.

Maerl

6.4.7. Maerl is one of the world's slowest growing plants (Birkett et al., 1998). Studies have measured growth rates from tenths of millimetres to one millimetre per year (Adey & McKibbin, 1970; Bosence & Wilson, 2003). The life span of individual plants of Lithothamnion glaciale has been estimated as 10-50 years (Adey & McKibbin, 1970). Spores can potentially disperse long distances although distances would be extremely limited if vegetative propagation was the key dispersal mechanism ( OSPAR, 2010).

6.4.8. Given the slow growth rates of maerl, individual plants and beds are slow to recover from damaging impacts. Their recovery potential has been characterised by OSPAR as 'poor' meaning that only partial recovery is likely within 10 years and full recovery may take up to 25 years (IMPACT, 1998). Maerl beds may never recover from severe damage such as bed removal, for example through dredging ( OSPAR, 2010, Hiscock et al., 2005).

6.4.9. Steller et al. (2003) found that the morphology of the maerl strongly influences the diversity of the species present. Any damage or removal of the maerl thalli would alter the complexity of the surface matrix thus reducing the interstitial space and complexity, and in turn the maerl's ability to provide habitat and shelter to various species.

Seagrasses

6.4.10. Zostera beds can undergo considerable annual and seasonal variation and the factors underpinning these changes are not always clear (Dale et al., 2007). Throughout the range intertidal populations are often annual and can undergo complete dieback in winter with recovery dependent on local seed supply (Holt et al., 1997). In perennial populations (lifespan over two years) die back of above ground parts is less significant and recovery is through vegetative growth. Zostera beds are also spatially dynamic, with advancing and leading edges causing changes in coverage. The beds expand either through vegetative growth from shooting rhizomes that have survived the winter, or sexually, by production of seed. Subtidal Z. marina beds in the UK are perennial and are believed to persist almost completely as a result of vegetative growth rather than by seed. Growth of individual plants occurs during the spring and summer. Recovery rates will therefore depend on supply of rhizomes. Given that fragmentation of beds can cause further losses, recovery may be slow, particularly in subtidal areas.

6.5. Seagrass regeneration

6.5.1. Recovery time of seagrasses after disturbance varies with seagrass species ( ABPmer, 2013). The slow recovery of Zostera populations since the 1920s to 30s outbreak of wasting disease suggests that, once lost, seagrass beds take considerable time to re-establish, if at all (Tillin et al., 2010). However, Phillips & Menez (1988) reported that displacement rhizomes and shoots can root and re-establish themselves if they settle on sediment long enough (cited in Huntington et al., 2006).

6.5.2. Zostera noltii, which is intertidal, can fill in gaps in seagrass meadows of 0.13 m ² in 1 month (Han et al., 2012 and preceding references therein). Disturbance size, disturbance intensity, sediment characteristics and seasonal time of disturbance are also likely to be influencing factors. Seagrasses can recover via lateral rhizome spread or via sexual reproduction and seed dispersal depending on location and species. The dispersal range of seagrass seeds is a very poorly studied aspect of their reproductive ecology, and robust estimates of dispersal events are only available for Z. marina populations, for which 95% of the seeds are retained within 30 m from the source.

6.5.3. Z. noltii is able to recover relatively quickly compared to other seagrass species (D'Avack et al., 2015; Tyler-Walters & Wilding, 2008a). However, potential recruitment of Z. noltii may be hampered by competition with infauna such as the ragworm Hediste diversicolor or lugworm Arenicola marina (Philippart, 1994; Hughes et al., 2000; cited in Tyler-Walters and Wilding, 2008a). Hughes et al. (2000) noted that H. diversicolor consumed leaves and seeds of Z. noltii by pulling them into their burrow, therefore reducing the survival of seedlings.

6.5.4. Cooke & McMath (2001) calculated the likely recovery potential of Z. marina in response to human maritime activities, based on the recruitment, recolonisation and regenerative characteristics of the species. On a scale of 1-100 (where 1 represented excellent recovery following disturbance and 100 represented no species recovery), the authors calculated that Z. marina had an intermediate recovery score of 49.

6.5.5. Recoverability of Z. marina will depend on recruitment from other populations where extraction occurs on a large scale across an entire bed. Although Z. marina seed dispersal may occur over large distances, high seedling mortality and seed predation may significantly reduce effective recruitment. Holt et al. (1997) suggested that recovery would take between 5-10 years, but in many cases would be longer.

6.5.6. Reed & Hovel (2006) found that removal of 90% of the substrate (which included seagrass plant material both above and below ground) in large 16 m ² plots resulted in a significant loss of diversity and abundance of the associated epifauna. However, in smaller plots, or with a lower level of substrate removal, there was no observed correlation between seagrass loss and reduction in density or diversity of epifaunal species.

6.6. Seagrass harvest methods

6.6.1. Recovery rates will also be influenced by the method of harvesting. Cutting leaves, either mechanically or by hand, will leave root and rhizome structures in place. Effects of cutting are therefore likely to be similar to those caused by grazing, whereby a seagrass bed would be expected to recover to pre-harvesting density within a year (Peterken and Conacher, 1997; Ganter, 2000).

6.6.2. Surface penetrating harvesting methods which disturb the below ground biomass of seagrasses, such as dredging, is likely to be more detrimental. Dredging can also have indirect detrimental effects by increasing suspended sediment (reducing light for photosynthesis) and elevating sedimentation (resulting in smothering). Such conditions may also lead to excessive growth of opportunistic epiphytic algal species potentially compromising the health and viability of seagrasses by overlying and smothering. Although seagrasses can potentially recover from this type of disturbance, recovery times are likely to be longer than those caused by cutting.

6.6.3. Seagrass beds may also be disturbed during harvesting activities that result in trampling of the substrate. These physical disturbances may lead to habitat loss and fragmentation (Reed and Hovel, 2006). A study on seagrass species in Puerto Rico found that changes to seagrass biomass as a result of trampling were inversely related to trampling intensity and duration (Eckrich and Holmquist, 2000). Substrate firmness was also found to modify trampling effects, with firmer substrates being less susceptible to damage than softer substrates.

6.7. The seagrass meadow habitat

6.7.1. Many species use seagrasses as nursery grounds and so the harvesting of seagrasses can have a negative effect on these species. However, a study by Heck et al. (2003) reports that it may not be the seagrass feature itself which is increasing survival and growth rate of juvenile species, but rather the structure of the habitat. Heck et al. (2003) report that there was very little difference between growth on seagrass habitats and other structured habitats that provide shelter.

Beach-casts

6.7.2. Evidence indicates that the removal of beach-casts will reduce biodiversity of this strandline habitat (Lavery et al., 1999; Dugan et al., 2003; Gilburn, 2012) and also the complexity of the trophic food web (Orr, 2013). Their removal also has the potential to change macrofaunal community structure and the prey availability for vertebrate species such as shorebirds (Dugan et al., 2003).

6.7.3. In Western Australia, cleaning the beach caused an immediate reduction in the biomass of macrophyte detritus and densities of epifauna and fish. Biomass at the cleaned beach returned to levels found in areas which had not been cleared (control beaches) within two months and it was concluded that there was no long term effect on sediment organic matter, density or richness values of benthic infauna (Lavery et al., 1999). Notably, although biomass richness recovered rapidly, the assemblage of species present was different in the cleaned and control beaches.

6.7.4. A food web model predicted that harvesting ( i.e. gathering) beach-cast kelps would also result in a proportional and immediate decline in primary consumers (Orr, 2013). The recovery time of the primary consumers was predicted to be 1-2 years independent of harvest intensity.

6.7.5. This food web model also predicted a decline in the numbers of shorebirds feeding on beach-cast kelps following gathering. The rate of recovery of shorebirds would be slow (2 to 60 years) and proportional to gathering intensity (Orr, 2013). Where more than 50% of the beach-cast material is removed, waders would reduce to less than 10% of their pre-harvest population and the recovery of these species increased from 13 years to 45 - 60 years (Orr, 2013). Similar results were reported for gulls. In order to allow shorebird populations to recover within a decade following the cessation of gathering, Orr (2013) suggests that no more than 30-40% of the beach-cast kelps should be gathered. Birds moving elsewhere as a result of the loss of beach-cast material would be regarded as a particular issue in designated areas or if protected birds species were being affected ( SNH, pers. comm.).

Opportunistic and Non-Native Species

6.7.6. The removal of native seaweeds could provide opportunity for the establishment of non-native seaweeds which could pose a threat to native species ( ABPmer, 2013). As non-native species are difficult to eradicate, their introduction may permanently change the character of a habitat ( OSPAR, 2009; ABPmer, 2013), having implications for those species which rely on seaweeds to provide habitat, shelter and food.

6.7.7. A large proportion of the large-scale variations in algal cover between areas in Denmark was found to be due to differences in water clarity and salinity (Krause-Jensen et al., 2007). This study reported that brackish waters, in particular, were vulnerable to an increase in opportunistic species. Further, a study in Norway found that the principal factors responsible for the replacement of Saccharina latissima by opportunistic and ephemeral filamentous algae in Skagerrak were wave and light exposure (Bekkby & Moy, 2011). Therefore, areas most vulnerable to the introduction of non-native seaweeds are likely to be brackish and sheltered waters as any changes in environmental conditions would be most predominant in these locations.