- Email: [email protected]
Restriction of Spartina anglica (C.E. Hubbard) marsh development by the infaunal polychaete Nereis diversicolor (O.F. Müller)
Estuarine, Coastal and Shelf Science 71 (2007) 202e209 www.elsevier.com/locate/ecss
Restriction of Spartina anglica (C.E. Hubbard) marsh development by the infaunal polychaete Nereis diversicolor (O.F. Mu¨ller) O.A.L. Paramor1, R.G. Hughes* School of Biological Sciences, Queen Mary and Westfield College, University of London, London, E1 4NS, UK Received 12 August 2005; accepted 25 July 2006 Available online 8 September 2006
Abstract Spartina anglica has been planted around the coast of SE England, and throughout the world, to help stabilise sediments and reduce wave erosion of sea defences. Recently the areas of S. anglica have declined on this coast, contributing to the overall loss of saltmarsh area, in contrast to the north and west coasts of the UK where expanding S. anglica marshes have had to be controlled to maintain mudflat feeding areas for birds. Field experiments were established in two old coastal realignment areas in the Blythe Estuary (Suffolk) and on open mudflats in the Blackwater Estuary (Essex), in which surface deposit feeding by Nereis diversicolor was prevented by laying porous mats on the mudflat surface. These mats promoted sediment deposition and quickly became buried. S. anglica colonised these infauna exclusion areas by growth of rhizomes from adjacent plants, to a significantly higher degree than in control areas. In one of the Blythe sites this occurred when the adjacent S. anglica marsh was retreating. These results support the conclusions of Paramor and Hughes (2004, 2005), that the infauna are responsible for much of the loss of saltmarsh in SE England, and that managed coastal realignment will not necessarily lead to saltmarsh creation. The experiments point to an alternative means of managing saltmarsh creation and sediment accretion on existing mudflats, with consequent benefits for coastal defence and conservation. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Spartina anglica; Nereis diversicolor; saltmarsh erosion; UK; Suffolk; Essex
1. Introduction A major change to some coastal ecosystems around the world has been the deliberate and accidental introduction of Spartina species to saltmarsh and mudflat habitats outside their natural distribution range. For example four species of Spartina have been introduced into San Francisco Bay (Ayres et al., 2004). Deliberate introductions of Spartina alterniflora and Spartina anglica, for example, have been to enhance coastal flood defence, as Spartina plants promote sediment accretion, stabilise sediments and reduce wave erosion of sea walls (Adam, 1990; Gray et al., 1991; Ayres et al., 2004; * Corresponding author. E-mail address: [email protected] (R.G. Hughes). 1 Current address: University of Liverpool, School of Biological Sciences, Bioscience Building, Crown Street, Liverpool, L69 7ZB, UK. 0272-7714/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ecss.2006.07.012
Chen et al., 2004). Another consequence of some introductions has been the creation of new species, as Spartina species readily hybridise. These hybrids often have a high fecundity and are fast growing, which increases their beneficial consequences. In the UK, S. anglica, a polyploid hybrid between S. alterniflora introduced from the USA and the native Spartina maritima, can colonise mudflats at lower elevations than the other native pioneer zone species (Gray et al., 1991). This led to an appreciation of its potential importance in reducing coastal erosion and from the 1920s S. anglica was planted on mudflats around the UK, especially on the subsiding coastline of SE England, and around the world (Gray et al., 1991). The invasive nature of some Spartina species may have detrimental consequences too; they can reduce the areas of mudflat that are feeding areas for birds (Goss-Custard and Moser, 1990; Percival et al., 1998; Chen et al., 2004), outcompete native species (Chen et al., 2004) and prevent the desired
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
outcome of coastal restoration schemes (Callaway, 2005). When controlling invasive Spartina becomes the priority it proves difficult and expensive (Frid et al., 1999; Taylor and Hastings, 2004) and may cause increased sediment erosion (Swales et al., 2005). The saltmarshes on the subsiding coast of SE England have been disappearing rapidly for the past half-century (Burd, 1992) with detrimental consequences for conservation and flood defence interests (see Hughes, 2004; Hughes and Paramor, 2004). These losses have been mainly of the pioneer zone vegetation, but erosion at the saltmarsh face and internal creek widening also has caused loss of vegetation from higher levels. These losses of vegetation have been attributed, in part, to bioturbation and herbivory by the infaunal polychaete Nereis diversicolor (Paramor and Hughes, 2004, 2005; Hughes and Paramor, 2004). The worms destabilise the sediments by surface deposit-feeding on microphytobenthos (Smith et al., 1996) and reduce colonisation of mudflats by eating the seeds and seedlings of annual pioneer zone species (Paramor and Hughes, 2004). This investigation complements the studies of Paramor and Hughes (2004, 2005) on annual pioneer zone species and tests the hypothesis that by excluding surface deposit feeding by the infauna (mostly N. diversicolor) in field experiments, perennial Spartina anglica can be stimulated to grow out onto mudflats. The effects of Nereis diversicolor may be particularly acute in SE England because they have a high vertical distribution range that overlaps that of saltmarsh vegetation, but seemingly not so in the north and west (Hughes and Paramor, 2004). Spartina anglica has been declining in area in the south and east of the UK (contributing to the overall loss of saltmarsh area), while expanding its range in the north and west of the UK (Charman, 1990; Gray et al., 1990), a geographical difference similar to that of saltmarshes generally. The causes of this dieback of S. anglica, and the dieback of Spartina elsewhere, are unknown but changes to sediment characteristics caused by the plants themselves may contribute to their decline (Gray et al., 1990, 1991; Ogburn and Alber, 2006). In this study the possibility that N. diversicolor may also contribute to this regional scale decline of S. anglica area is examined. Spartina anglica can colonise mudflats from seeds or fragments of plants and once established may spread laterally by vegetative growth of rhizomes, often leading to sudden rapid increases in coverage (Gray et al., 1991). Hughes et al. (2000) demonstrated that N. diversicolor restricted similar vegetative colonisation of mudflats by Zostera noltii and observed N. diversicolor burying S. anglica seeds, by grasping them and dragging them into their burrow, often to depths greater than 4 cm which reduces their germination success (Groenendijk, 1986). Restoration of saltmarshes in SE England is an urgent conservation and coastal defence priority (Hughes and Paramor, 2004). Saltmarsh has been declared a Biodiversity Action Plan habitat (United Kingdom Biodiversity Group, 1999), and the UK Government is committed to restoring and maintaining the area of saltmarsh present in 1992. The main strategy is to use managed realignment, where in selected areas of
203
the coast breaching the sea defences will allow saltmarsh to develop on agricultural land which has become intertidal. This policy is controversial (Hughes and Paramor, 2004; Morris et al., 2004;Wolters et al., 2005) because it is based on the assumption that saltmarsh loss in SE England is due to sea level rise leading to coastal squeeze, which Hughes and Paramor (2004) argued was a false assumption. Further, Paramor and Hughes (2005) argued that low-lying realignment sites would not develop saltmarsh vegetation because the infauna that colonise accreting sediments would prevent it. This may be why older low-lying unmanaged realignment areas remain as unvegetated mudflats. These conclusions were based on studies with annual pioneer zone species in recent managed realignment areas. In this investigation with Spartina anglica one of the study areas includes two realignment sites on the Blythe estuary, which after more than 50 years are still mainly mudflats. If these experiments successfully promote colonisation of mudflats by S. anglica this would reinforce the conclusion of Paramor and Hughes (2005) that managed realignment cannot be relied upon to create sufficient areas of new saltmarsh, and, more importantly, would point to a cheap practical alternative management technique for increasing saltmarsh area without realignment. For infaunal exclusion to be of significant management potential it would have to work with S. anglica, because with the loss of so much of the annual pioneer zone vegetation in SE England S. anglica is often the only pioneer zone plant species remaining. 2. Methods The experiments were conducted in the Blythe Estuary, Suffolk, and at Maldon, Essex, at the head of the Blackwater Estuary (Fig. 1). In the Blythe Estuary infauna exclusion was
Fig. 1. The locations of the experimental sites on the Blythe Estuary (B2, B3) and at Maldon on the Blackwater Estuary (M).
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
The experiments at Maldon (TM 858 075) were on a mudflat in a wave-sheltered area adjacent to patches of Spartina anglica. In March 2000 five (0.25 m2) mulchmats were pinned to the sediment adjacent to patches of S. anglica and five similar areas of sediment were identified as controls. After 5 months (in August) the depth of sediment accreted above the mats was measured and the abundance of the invertebrates and plants in the exclusion and control areas assessed, as described above. 3. Results 3.1. Blythe Estuary In both sites in the Blythe estuary sediment had accreted on all the mats in 6 months, with more at site B2 than B3 and more over the nets than the mulchmats, with up to 3 cm over the nets at B2 (Fig. 2). The mats were still at the level of the surrounding sediment indicating that little or no accretion had occurred there. In only 6 months Spartina anglica had colonised the mats at both sites (Fig. 3), and although the mean abundances were low they were significantly greater than in the control areas, where no plants were present, (t-test, t ¼ 2.4, d.f. ¼ 8, p ¼ 0.04 for B2 and t ¼ 3.21, d.f. ¼ 8, p ¼ 0.012 for B3). These plants had grown up through the nets and Site B2 3
2
cm
achieved by placing two types of mat, 28 gauge cotton netting and mulchmat, on the sediment to prevent the infauna (mostly Nereis) from reaching the surface to deposit-feed. At Maldon only mulchmat was used (see Paramor and Hughes (2004) for a rationale of the approach and a full description of the mats). Morris et al. (2004) and Wolters et al. (2005) criticised the experimental design as there was no procedural control to account for the direct effects of the mats on deposition and resuspension of sediment. However, since the hypothesis was simply that placing mats on the surface would lead to sediment accretion and saltmarsh development, and did not seek to separate the biological and physical effects within the experiment, no such procedural control was necessary. Moreover, as the mats promoted rapid sedimentation and were quickly buried the physical effects of the mats on sediment accretion were only temporary. Three experiments were established in the Blythe Estuary at the sites B1, B2 and B3 of French et al. (2000), who gave maps and detailed descriptions of these areas. The first two were in the Bulcamp mudflats, a large realignment area formed accidentally in 1940 that has developed into mudflats with only a narrow peripheral zone of Spartina anglica in places. The first experiment (B1) was in the most wave-exposed area at the eastern end of the mudflats, but as the highly mobile sandy sediment contained few infauna this experiment was discontinued. The second (B2) was in a wave-sheltered location (National Grid Reference TM 478 766) where the sediment was a soft mud. The third experiment (B3) was in a smaller realignment area near Blackshore Mill (TM 494 759) created in 1953 and which has also remained mainly as mudflat with a peripheral zone of S. anglica. Both experimental areas were at approximately mean high water neap tide level (MHWNTL). In December 1998 five mulchmats (0.25 m2) and five nets (0.25 m2) were pinned to the sediment close to the edge of the Spartina anglica marsh at B2 and B3. A similar area of mudflat close to each mat was designated as a control at the outset. This study was short-term, because of access restrictions in the Blythe Estuary, and in June 1999 (the only opportunity to gather data), the depths of accreted sediment over the mats were measured using a rule. The elevation of the sediment in the control areas was compared to the level of the mats. Any plants in each experimental and control area were identified and counted. Small areas of each mat (approximately 5 5 cm) and the overlying sediment were collected to assess the density of invertebrates. In previous experiments the sediment over the mats became colonised by small invertebrates, possibly as it was a refuge from predation or disturbance from larger Nereis diversicolor, and these could have affected plant colonisation (Paramor and Hughes, 2005). A core sample (7.5 cm diameter, 20 cm depth) was taken from each control area at the same time. The invertebrates were removed from these sediment samples, as described by Paramor and Hughes (2004), identified and counted. Samples were not taken from underneath the mats to avoid disturbing them, but in similar studies macrofauna were found only rarely under mats (RGH, personal observation).
1
0
mulch
net
Site B3 3
2
cm
204
1
0
mulch
net
Fig. 2. The mean depths (þs.e., n ¼ 5) of sediment accreted in six months over the two types of mat laid on the sediment at sites B2 and B3 in the Blythe Estuary.
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Site B2 6
no.m-2
4
2
0
control
mulch
net
Site B3 6
no.m-2
4
2
205
30 months, while the edge of the established S. anglica marsh had retreated unexpectedly (Fig. 4). The densities of the macrofauna found in the sediment over the mats and in the adjacent control areas are shown in Fig. 5. The densities of Nereis diversicolor in the sediment over the mats at B2 were significantly less than in the control areas, where a mean density of over 2500 m2 was recorded (t-test t ¼ 4.77, d.f. ¼ 8, p ¼ 0.02 and t ¼ 4.15, d.f. ¼ 8, p ¼ 0.2 for mulchmats and nets, respectively). The N. diversicolor over the mats were all small (less than 2 cm in length), in contrast to the worms in the control areas which were up to 10 cm long, a difference in size frequency similar to that recorded elsewhere in June by Paramor and Hughes (2004). At B3 the mean densities of worms in the control areas were much lower, and while no N. diversicolor were found over the mulchmats, small worms (less than 2 cm in length) were present over the nets at a mean density similar to that of the larger worms in the control areas. The mudsnail Hydrobia ulvae and the bivalve Macoma balthica were also present at both sites but with higher densities at B3. In neither case were the densities over the mats and control areas significantly different. Corophium volutator, Terebellides stroemi, Ophelia rathkei and Nephtys caeca were also present at B3, but not at B2. For none of these species were the abundances in the experimental and control areas consistently different. 3.2. Blackwater Estuary.
0
control
mulch
net
Fig. 3. The mean density (þs.e., n ¼ 5) of Spartina anglica plants in the control areas and growing through the mats after six months in the Blythe Estuary experiments.
mulchmats from rhizomes that had extended laterally from the adjacent plants. In 2001 site B3 was revisited briefly and although no data could be collected significant growth of S. anglica in the exclusion areas had occurred over the previous
At Maldon the mean depth of sediment accreted on the mulchmats after 5 months was 2.75 cm (0.41 s.e.). These mats were still at the level of the surrounding sediment indicating that little or no deposition had occurred there. In just 5 months these mats were colonised by Spartina anglica plants at a mean density of over 40 m2, a significantly higher number than in the control areas (t ¼ 3.72, d.f. ¼ 8, p ¼ 0.006) (Fig. 6). The control areas had a significantly higher mean density of Nereis diversicolor than over the mats (t ¼ 2.89, d.f. ¼ 8, p ¼ 0.02) (Fig. 6), and these worms were larger
Fig. 4. An example of the development of Spartina anglica marsh over mulchmat (D) and net (E) after 18 months, (and further out, into the area disturbed by treading when samples were collected in June 2000) in comparison to the retreat of the adjacent marsh (A, B, C).
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Site B2
Site B3 3000
no.m-2
3000
2000 1000 0
control
mulch
2000 1000 0
net
1500
6000
4000
4000
1000 500
control
mulch
2000
250 200 150 100 50 0
mulch
control
mulch
net
control
mulch
net
9000
no.m-2
3000 control
mulch
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
2000
1000 800 600 400 200 0
250 200 150 100 50 0
6000 3000 0
net
600
600 400 200 0
net
9000
6000
0
mulch
500
0
net
no.m-2
1000 800 600 400 200 0
control
no.m-2
0
control
1000 0
net
no.m-2
0
no.m-2
Terebellides stroemi no.m-2 Ophelia rathkei no.m-2 Nepthys caeca no.m-2
no.m-2
6000
Corophium volutator no.m-2
Hydrobia ulvae no.m-2
2000
1500
2000
Macoma balthica no.m-2
Nereis diversicolor no.m-2
206
control
mulch
net
400 200 0
Fig. 5. The mean densities (þs.e., n ¼ 5) of the invertebrate infauna in the control areas and in the sediment accreted over the experimental mats after six months at the two Blythe Estuary experimental sites.
than those over the mats, as in the Blythe (see above). Hydrobia ulvae were also present at similar densities, of around 600 m2, over the mats and in the control areas. 4. Discussion The infauna exclusion mats at all three locations stimulated significant deposition of sediment and colonisation by Spartina
anglica. Initially, the sediment accretion would have been due to the mats directly enhancing sediment deposition and reducing resuspension, but this effect would have been only temporary because they were rapidly buried. The continued deposition of sediment on the buried mats can be attributed to exclusion of deposit-feeding, mostly by Nereis diversicolor, since no accretion was observed in the adjacent control areas, which remained at the same level as the mats. In the Blythe,
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Spartina anglica 60
no.m-2
40
20
0
control
mulch
Nereis diversicolor 1200 1000
no.m-2
800 600 400 200 0
control
mulch
Fig. 6. The mean densities (þs.e., n ¼ 5) of the Nereis diversicolor and Spartina anglica in the control areas and in the sediment accreted over the mulchmats after five months at Maldon, Blackwater Estuary.
at Blackshore Mill (the location of B3) French et al. (2000) measured an increase in sediment elevation from September to March but a compensating decrease in the following 6 months. In the Bulcamp Marshes (the location of B2) there was no overall change in the sediment elevations and only 13 cm of sediment had accreted in 55 years (French et al., 2000). In this study the nets at B2 accreted 3 cm in only 6 months. The high rates of sediment accretion in the infauna exclusion areas support the conclusion of Paramor and Hughes (2004, 2005) that N. diversicolor destabilise sediments, by bioturbation and eating the microphytobenthos that help stabilize sediments with their mucus-like secretions, and lower the elevations at which the sediments achieve dynamic stability between the opposing processes of deposition and resuspension. In previous similar exclusion experiments relatively high densities of small Nereis diversicolor and other species were found in the sediment over the mats, perhaps because these areas were a refuge from predation or disturbance by large N. diversicolor (Paramor and Hughes, 2005). A similar result was not found in these experiments perhaps because of their short duration. However, in the control areas at B3, where there was a relatively low density of N. diversicolor, six other invertebrate species were present at relatively high densities,
207
while at B2, where there was a high density of N. diversicolor, only low densities of Hydrobia ulvae and Macoma balthica were recorded. These differences may be interpreted as further evidence that N. diversicolor adversely affects other infaunal species. More directed experiments are required to test this hypothesis. In just a few months the infauna exclusion areas were colonised by Spartina anglica shoots at densities significantly greater than in the control areas, and at B3 even when the adjacent S. anglica bed was dying back. That some S. anglica colonised the control areas at Maldon indicates that S. anglica were developing onto the mudflats, but no such long-term expansion of the edge of the S. anglica marsh has been apparent since these experiments were conducted (RGH, personal observation). The successful colonisation of the exclusion areas by S. anglica, by shoots that had grown upward from rhizomes that had extended underneath the mats from the adjacent plants, indicates that under normal conditions such development of the S. anglica marsh is prevented or restricted by Nereis diversicolor. This may be because the worms disturb or damage developing rhizomes or roots, as was shown with Zostera noltii (Hughes et al., 2000). Emmerson (2000) found that small S. anglica plants transplanted into N. diversicolor exclusion areas were slightly, but significantly, heavier than those transplanted into control areas, and their roots were longer. Spartina anglica, and S. alterniflora in the USA, have suffered from die-back in some areas, leaving sediments and sea walls more open to erosion. The causes of the die-back are uncertain but may include pathogens, grazing or increased waterlogging of the sediment in the centre of Spartina stands. Low oxygen concentrations in waterlogged sediments may reduce meristem growth and lead to increased concentrations of acetylene and sulphide (Gray et al., 1991), or ammonium (Ogburn and Alber, 2006), all of which are toxic to plants. The observation at B3 that the S. anglica marsh was regressing, when the exclusion experiments promoted development of S. anglica, indicates that interactions with infauna may also be a contributory factor to S. anglica dieback, particularly in the south and east of the UK, where Nereis diversicolor has a high abundance high in the intertidal zone (Paramor and Hughes, 2004) and where the loss of S. anglica in the UK has been the greatest. Levin et al. (2006) reported that an invasion of Spartina in San Francisco Bay resulted in a ‘‘trophic shift’’ where deposit feeding polychaetes (including nereids) that consumed Spartina detritus did not decrease in abundance in invaded areas, whereas other invertebrate taxa, including amphipods and bivalves, became less abundant. A similar shift in the infaunal invertebrate community, with increased polychaete dominance, has apparently occurred in SE England, and on the eastern shores of the Southern North Sea, for unknown reasons (Hughes, 1999). The results of Levin et al. leads to the intriguing hypothesis that the increase in abundance of Nereis diversicolor in SE England may be due, in part, to the increased supply of detrital food from the extensive S. anglica plantings (and their
208
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
subsequent spread) in the first half of the last century. Further, a positive feedback may have been established more recently, where N. diversicolor have additionally benefited from the detritus produced by the Spartina dieback which, this study indicates, may be accelerated by these polychaetes. This is in addition to the loss of other saltmarsh vegetation that they cause (Paramor and Hughes, 2004). A study of the food sources of the high density populations of N. diversicolor in the upper intertidal areas in SE England is required to test this hypothesis. The UK agencies responsible for saltmarsh conservation and restoration, The Environment Agency and English Nature, rely on managed realignment as the main method by which the UK is to try to achieve is international obligations regarding saltmarsh restoration (Morris et al., 2004). The results presented here from the two old realignment areas in the Blythe Estuary extend and support the conclusion of Paramor and Hughes (2005), that infaunal invertebrates will colonise the accreting sediment in low-lying managed realignment sites and will restrict or prevent saltmarsh development. This together with the insufficient availability of land for realignment (Covey and Laffoley, 2002), is why managed realignment is not the means by which the UK will be able to restore sufficient saltmarsh area to meet its obligations under the Biodiversity Action Plan. Morris et al. (2004) disputed this conclusion, but their arguments were flawed and the total area of vegetated saltmarsh created in managed realignment sites since 1995 in SE England is about the same as just 3 years losses (Hughes, in press). Morris et al. (2004) suggested that the lack of vegetation in the Blythe realignment areas after more than 50 years was due to ‘‘the physical profile of the estuary’’ but presented no supporting evidence. The results presented here show this is not correct. These experiments show that managed realignment is not necessary for saltmarsh creation, since the stationary or retreating edges of Spartina anglica marshes can be extended to seaward on existing mudflats. This result also confirms the conclusion of Hughes and Paramor (2004) that the lower limit of saltmarshes in SE England is not determined by sea level, and is, therefore, independent of rising sea level. Consequently the regional-scale decline in saltmarsh cannot be explained by coastal squeeze and managed realignment is not necessary for saltmarsh creation (Hughes and Paramor, 2004). Extending these experiments in scale and geographic range could be a simple and cheap means of managing development of pioneer zone vegetation and sediment accretion, replacing some of the eroding vegetation with economic benefits envisaged by those that originally transplanted S. anglica around this coast. S. anglica is perennial, relatively large, and morphologically more complex than the annual pioneer zone species. Consequently the benefits of increasing S. anglica marsh area, reduced wave impact on existing sea walls and ecological benefits associated with the high primary productivity of S. anglica (Long et al., 1990) would be greater. Deliberately extending S. anglica marshes onto mudflats might seem anathema to those in regions where such growth of
Spartina marshes is an ecological problem (see Section 1) but in SE England saltmarsh conservation would be considered more appropriate, given the continuing high rates of vegetation loss. References Adam, P., 1990. Saltmarsh Ecology. Cambridge University Press, Cambridge UK, 461 pp. Ayres, D.R., Smith, D.L., Zaremba, K., Klohr, S., Strong, D.R., 2004. Spread of exotic cordgasses and hybrids (Spartina sp.) in the tidal marshes of San Francisco Bay California, USA. Biological Invasions 6, 221e231. Burd, F., 1992. Erosion and vegetation change on the saltmarshes of Essex and north Kent between 1973 and 1988. In: Research and Survey in Nature Conservation 42. Nature Conservancy Council, Peterborough. Callaway, J.C., 2005. The challenge of restoring functioning salt marsh ecosystems. Journal of Coastal Research 21, 24e36 (special issue 40). Charman, K., 1990. The current status of Spartina anglica in Britain. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 11e14. Chen, Z.Y., Li, B., Zhong, Y., Chen, J.K., 2004. Local competitive effects of introduced Spartina alterniflora on Scirpus mariqueter at Dongtan of Chongming Island, the Yangtze River estuary and their potential ecological consequences. Hydrobiologia 528, 99e106. Covey, R., Laffoley, D.d’A., 2002. Maritime State of Nature Report for England; Getting Onto an Even Keel. English Nature, Peterborough, UK. Emmerson, M., 2000. Remedial habitat creation: does Nereis diversicolor play a confounding role in the colonisation and establishment of the pioneering saltmarsh plant Spartina anglica? Helgoland Marine Research 54, 110e116. French, C.E., French, J.R., Clifford, N.J., Watson, C.J., 2000. Sedimentationerosion dynamics of abandoned reclamations. Continental Shelf Research 20, 1711e1733. Frid, C.L.J., Chandrasekara, W.U., Davey, P., 1999. The restoration of mud flats invaded by common cord-grass (Spartina anglica, CE Hubbard) using mechanical disturbance and its effects on the macrobenthic fauna. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 47e61. Goss-Custard, J.D., Moser, M.E., 1990. Changes in the numbers of dunlin (Calidris alpina) in British estuaries in relation to changes in the abundance of Spartina. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication No. 2. HMSO, London, UK, pp. 69e71. Gray, A.J., Benham, P.E.M., Raybould, A.F., 1990. Spartina anglica e the evolutionary and ecological background. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A RESEARCH REVIEW. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 5e10. Gray, A.J., Marshall, D.F., Raybould, A.F., 1991. A century of evolution in Spartina anglica. Advances in Ecological Research 21, 1e62. Groenendijk, A.M., 1986. Establishment of a Spartina anglica population on a tidal mudflat: a field experiment. Journal of Environmental Management 22, 1e12. Hughes, R.G., 1999. Saltmarsh erosion and management of saltmarsh restoration; the effects of infaunal invertebrates. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 83e95. Hughes, R.G., 2004. Climate change and loss of saltmarshes; consequences for birds. Ibis 146 (Suppl. 1), 21e28. Hughes, R.G. On the loss of saltmarshes in south-east England and the utility of managed realignment in their restoration. Journal of Applied Ecology, in press. Hughes, R.G., Lloyd, D., Ball, L., Emson, D., 2000. The effects of the polychaete Nereis diversicolor on the distribution and transplanting success of Zostera noltii. Helgoland Marine Research 54, 129e136. Hughes, R.G., Paramor, O.A.L., 2004. On the loss of saltmarshes in south-east England and methods for their restoration. Journal of Applied Ecology 41, 449e463.
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209 Levin, L.A., Neira, C., Grosholz, E.D., 2006. Invasive cordgrass modifies wetland trophic function. Ecology 87, 419e432. Long, S.P., Dunn, R., Jackson, D., Othman, S.B., Yaakub, M.H., 1990. The primary productivity of Spartina anglica on an East Anglian estuary. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 34e38. Morris, R.K.A., Reach, I.S., Duffy, M.J., Collins, T.S., Leafe, R.N., 2004. On the loss of saltmarshes in south-east England and the relationship with Nereis diversicolor. Journal of Applied Ecology 41, 787e791. Ogburn, M.B., Alber, M., 2006. An investigation of saltmarsh dieback in Georgia using field transplants. Estuaries and Coasts 29, 54e62. Paramor, O.A.L., Hughes, R.G., 2004. The effects of bioturbation and herbivory by the polychaete Nereis diversicolor on loss of saltmarsh in south-east England. Journal of Applied Ecology 41, 440e448. Paramor, O.A.L., Hughes, R.G., 2005. The effects of the invertebrate infauna on the early colonisation of managed realignment areas by saltmarsh plants in SE England. Marine Ecology Progress Series 303, 61e71.
209
Percival, S.M., Sutherland, W.J., Evans, P.R., 1998. Intertidal habitat loss and wildfowl numbers: applications of a spatial depletion model. Journal of Applied Ecology 35, 57e63. Smith, D., Hughes, R.G., Cox, E.J., 1996. Predation of epipelic diatoms by the amphipod Corophium volutator and the polychaete Nereis diversicolor. Marine Ecology Progress Series 145, 53e61. Swales, A., Ovenden, R., MacDonald, I.T., Lohrer, A.M., Burt, K.L., 2005. Sediment remobilisation from decomposing cordgrass (Spartina anglica) patches on a wave-exposed intertidal flat. New Zealand Journal of Marine and Freshwater Research 39, 1305e1319. Taylor, C.M., Hastings, A., 2004. Finding optimal control strategies for invasive species: a density-structured model for Spartina alterniflora. Journal of Applied Ecology 41, 1049e1057. United Kingdom Biodiversity Group, 1999. Maritime Species and Habitats, vol. V. Tranche 2 Action Plans. English Nature for the UK Biodiversity Action Group, Peterborough. Wolters, M., Bakker, J.P., Bertness, M.D., Jefferies, R.L., Mo¨ller, I., 2005. Salt-marsh erosion and restoration in south-east England: squeezing the evidence requires realignment. Journal of Applied Ecology 42, 844e851.
Restriction of Spartina anglica (C.E. Hubbard) marsh development by the infaunal polychaete Nereis diversicolor (O.F. Mu¨ller) O.A.L. Paramor1, R.G. Hughes* School of Biological Sciences, Queen Mary and Westfield College, University of London, London, E1 4NS, UK Received 12 August 2005; accepted 25 July 2006 Available online 8 September 2006
Abstract Spartina anglica has been planted around the coast of SE England, and throughout the world, to help stabilise sediments and reduce wave erosion of sea defences. Recently the areas of S. anglica have declined on this coast, contributing to the overall loss of saltmarsh area, in contrast to the north and west coasts of the UK where expanding S. anglica marshes have had to be controlled to maintain mudflat feeding areas for birds. Field experiments were established in two old coastal realignment areas in the Blythe Estuary (Suffolk) and on open mudflats in the Blackwater Estuary (Essex), in which surface deposit feeding by Nereis diversicolor was prevented by laying porous mats on the mudflat surface. These mats promoted sediment deposition and quickly became buried. S. anglica colonised these infauna exclusion areas by growth of rhizomes from adjacent plants, to a significantly higher degree than in control areas. In one of the Blythe sites this occurred when the adjacent S. anglica marsh was retreating. These results support the conclusions of Paramor and Hughes (2004, 2005), that the infauna are responsible for much of the loss of saltmarsh in SE England, and that managed coastal realignment will not necessarily lead to saltmarsh creation. The experiments point to an alternative means of managing saltmarsh creation and sediment accretion on existing mudflats, with consequent benefits for coastal defence and conservation. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Spartina anglica; Nereis diversicolor; saltmarsh erosion; UK; Suffolk; Essex
1. Introduction A major change to some coastal ecosystems around the world has been the deliberate and accidental introduction of Spartina species to saltmarsh and mudflat habitats outside their natural distribution range. For example four species of Spartina have been introduced into San Francisco Bay (Ayres et al., 2004). Deliberate introductions of Spartina alterniflora and Spartina anglica, for example, have been to enhance coastal flood defence, as Spartina plants promote sediment accretion, stabilise sediments and reduce wave erosion of sea walls (Adam, 1990; Gray et al., 1991; Ayres et al., 2004; * Corresponding author. E-mail address: [email protected] (R.G. Hughes). 1 Current address: University of Liverpool, School of Biological Sciences, Bioscience Building, Crown Street, Liverpool, L69 7ZB, UK. 0272-7714/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ecss.2006.07.012
Chen et al., 2004). Another consequence of some introductions has been the creation of new species, as Spartina species readily hybridise. These hybrids often have a high fecundity and are fast growing, which increases their beneficial consequences. In the UK, S. anglica, a polyploid hybrid between S. alterniflora introduced from the USA and the native Spartina maritima, can colonise mudflats at lower elevations than the other native pioneer zone species (Gray et al., 1991). This led to an appreciation of its potential importance in reducing coastal erosion and from the 1920s S. anglica was planted on mudflats around the UK, especially on the subsiding coastline of SE England, and around the world (Gray et al., 1991). The invasive nature of some Spartina species may have detrimental consequences too; they can reduce the areas of mudflat that are feeding areas for birds (Goss-Custard and Moser, 1990; Percival et al., 1998; Chen et al., 2004), outcompete native species (Chen et al., 2004) and prevent the desired
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
outcome of coastal restoration schemes (Callaway, 2005). When controlling invasive Spartina becomes the priority it proves difficult and expensive (Frid et al., 1999; Taylor and Hastings, 2004) and may cause increased sediment erosion (Swales et al., 2005). The saltmarshes on the subsiding coast of SE England have been disappearing rapidly for the past half-century (Burd, 1992) with detrimental consequences for conservation and flood defence interests (see Hughes, 2004; Hughes and Paramor, 2004). These losses have been mainly of the pioneer zone vegetation, but erosion at the saltmarsh face and internal creek widening also has caused loss of vegetation from higher levels. These losses of vegetation have been attributed, in part, to bioturbation and herbivory by the infaunal polychaete Nereis diversicolor (Paramor and Hughes, 2004, 2005; Hughes and Paramor, 2004). The worms destabilise the sediments by surface deposit-feeding on microphytobenthos (Smith et al., 1996) and reduce colonisation of mudflats by eating the seeds and seedlings of annual pioneer zone species (Paramor and Hughes, 2004). This investigation complements the studies of Paramor and Hughes (2004, 2005) on annual pioneer zone species and tests the hypothesis that by excluding surface deposit feeding by the infauna (mostly N. diversicolor) in field experiments, perennial Spartina anglica can be stimulated to grow out onto mudflats. The effects of Nereis diversicolor may be particularly acute in SE England because they have a high vertical distribution range that overlaps that of saltmarsh vegetation, but seemingly not so in the north and west (Hughes and Paramor, 2004). Spartina anglica has been declining in area in the south and east of the UK (contributing to the overall loss of saltmarsh area), while expanding its range in the north and west of the UK (Charman, 1990; Gray et al., 1990), a geographical difference similar to that of saltmarshes generally. The causes of this dieback of S. anglica, and the dieback of Spartina elsewhere, are unknown but changes to sediment characteristics caused by the plants themselves may contribute to their decline (Gray et al., 1990, 1991; Ogburn and Alber, 2006). In this study the possibility that N. diversicolor may also contribute to this regional scale decline of S. anglica area is examined. Spartina anglica can colonise mudflats from seeds or fragments of plants and once established may spread laterally by vegetative growth of rhizomes, often leading to sudden rapid increases in coverage (Gray et al., 1991). Hughes et al. (2000) demonstrated that N. diversicolor restricted similar vegetative colonisation of mudflats by Zostera noltii and observed N. diversicolor burying S. anglica seeds, by grasping them and dragging them into their burrow, often to depths greater than 4 cm which reduces their germination success (Groenendijk, 1986). Restoration of saltmarshes in SE England is an urgent conservation and coastal defence priority (Hughes and Paramor, 2004). Saltmarsh has been declared a Biodiversity Action Plan habitat (United Kingdom Biodiversity Group, 1999), and the UK Government is committed to restoring and maintaining the area of saltmarsh present in 1992. The main strategy is to use managed realignment, where in selected areas of
203
the coast breaching the sea defences will allow saltmarsh to develop on agricultural land which has become intertidal. This policy is controversial (Hughes and Paramor, 2004; Morris et al., 2004;Wolters et al., 2005) because it is based on the assumption that saltmarsh loss in SE England is due to sea level rise leading to coastal squeeze, which Hughes and Paramor (2004) argued was a false assumption. Further, Paramor and Hughes (2005) argued that low-lying realignment sites would not develop saltmarsh vegetation because the infauna that colonise accreting sediments would prevent it. This may be why older low-lying unmanaged realignment areas remain as unvegetated mudflats. These conclusions were based on studies with annual pioneer zone species in recent managed realignment areas. In this investigation with Spartina anglica one of the study areas includes two realignment sites on the Blythe estuary, which after more than 50 years are still mainly mudflats. If these experiments successfully promote colonisation of mudflats by S. anglica this would reinforce the conclusion of Paramor and Hughes (2005) that managed realignment cannot be relied upon to create sufficient areas of new saltmarsh, and, more importantly, would point to a cheap practical alternative management technique for increasing saltmarsh area without realignment. For infaunal exclusion to be of significant management potential it would have to work with S. anglica, because with the loss of so much of the annual pioneer zone vegetation in SE England S. anglica is often the only pioneer zone plant species remaining. 2. Methods The experiments were conducted in the Blythe Estuary, Suffolk, and at Maldon, Essex, at the head of the Blackwater Estuary (Fig. 1). In the Blythe Estuary infauna exclusion was
Fig. 1. The locations of the experimental sites on the Blythe Estuary (B2, B3) and at Maldon on the Blackwater Estuary (M).
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
The experiments at Maldon (TM 858 075) were on a mudflat in a wave-sheltered area adjacent to patches of Spartina anglica. In March 2000 five (0.25 m2) mulchmats were pinned to the sediment adjacent to patches of S. anglica and five similar areas of sediment were identified as controls. After 5 months (in August) the depth of sediment accreted above the mats was measured and the abundance of the invertebrates and plants in the exclusion and control areas assessed, as described above. 3. Results 3.1. Blythe Estuary In both sites in the Blythe estuary sediment had accreted on all the mats in 6 months, with more at site B2 than B3 and more over the nets than the mulchmats, with up to 3 cm over the nets at B2 (Fig. 2). The mats were still at the level of the surrounding sediment indicating that little or no accretion had occurred there. In only 6 months Spartina anglica had colonised the mats at both sites (Fig. 3), and although the mean abundances were low they were significantly greater than in the control areas, where no plants were present, (t-test, t ¼ 2.4, d.f. ¼ 8, p ¼ 0.04 for B2 and t ¼ 3.21, d.f. ¼ 8, p ¼ 0.012 for B3). These plants had grown up through the nets and Site B2 3
2
cm
achieved by placing two types of mat, 28 gauge cotton netting and mulchmat, on the sediment to prevent the infauna (mostly Nereis) from reaching the surface to deposit-feed. At Maldon only mulchmat was used (see Paramor and Hughes (2004) for a rationale of the approach and a full description of the mats). Morris et al. (2004) and Wolters et al. (2005) criticised the experimental design as there was no procedural control to account for the direct effects of the mats on deposition and resuspension of sediment. However, since the hypothesis was simply that placing mats on the surface would lead to sediment accretion and saltmarsh development, and did not seek to separate the biological and physical effects within the experiment, no such procedural control was necessary. Moreover, as the mats promoted rapid sedimentation and were quickly buried the physical effects of the mats on sediment accretion were only temporary. Three experiments were established in the Blythe Estuary at the sites B1, B2 and B3 of French et al. (2000), who gave maps and detailed descriptions of these areas. The first two were in the Bulcamp mudflats, a large realignment area formed accidentally in 1940 that has developed into mudflats with only a narrow peripheral zone of Spartina anglica in places. The first experiment (B1) was in the most wave-exposed area at the eastern end of the mudflats, but as the highly mobile sandy sediment contained few infauna this experiment was discontinued. The second (B2) was in a wave-sheltered location (National Grid Reference TM 478 766) where the sediment was a soft mud. The third experiment (B3) was in a smaller realignment area near Blackshore Mill (TM 494 759) created in 1953 and which has also remained mainly as mudflat with a peripheral zone of S. anglica. Both experimental areas were at approximately mean high water neap tide level (MHWNTL). In December 1998 five mulchmats (0.25 m2) and five nets (0.25 m2) were pinned to the sediment close to the edge of the Spartina anglica marsh at B2 and B3. A similar area of mudflat close to each mat was designated as a control at the outset. This study was short-term, because of access restrictions in the Blythe Estuary, and in June 1999 (the only opportunity to gather data), the depths of accreted sediment over the mats were measured using a rule. The elevation of the sediment in the control areas was compared to the level of the mats. Any plants in each experimental and control area were identified and counted. Small areas of each mat (approximately 5 5 cm) and the overlying sediment were collected to assess the density of invertebrates. In previous experiments the sediment over the mats became colonised by small invertebrates, possibly as it was a refuge from predation or disturbance from larger Nereis diversicolor, and these could have affected plant colonisation (Paramor and Hughes, 2005). A core sample (7.5 cm diameter, 20 cm depth) was taken from each control area at the same time. The invertebrates were removed from these sediment samples, as described by Paramor and Hughes (2004), identified and counted. Samples were not taken from underneath the mats to avoid disturbing them, but in similar studies macrofauna were found only rarely under mats (RGH, personal observation).
1
0
mulch
net
Site B3 3
2
cm
204
1
0
mulch
net
Fig. 2. The mean depths (þs.e., n ¼ 5) of sediment accreted in six months over the two types of mat laid on the sediment at sites B2 and B3 in the Blythe Estuary.
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Site B2 6
no.m-2
4
2
0
control
mulch
net
Site B3 6
no.m-2
4
2
205
30 months, while the edge of the established S. anglica marsh had retreated unexpectedly (Fig. 4). The densities of the macrofauna found in the sediment over the mats and in the adjacent control areas are shown in Fig. 5. The densities of Nereis diversicolor in the sediment over the mats at B2 were significantly less than in the control areas, where a mean density of over 2500 m2 was recorded (t-test t ¼ 4.77, d.f. ¼ 8, p ¼ 0.02 and t ¼ 4.15, d.f. ¼ 8, p ¼ 0.2 for mulchmats and nets, respectively). The N. diversicolor over the mats were all small (less than 2 cm in length), in contrast to the worms in the control areas which were up to 10 cm long, a difference in size frequency similar to that recorded elsewhere in June by Paramor and Hughes (2004). At B3 the mean densities of worms in the control areas were much lower, and while no N. diversicolor were found over the mulchmats, small worms (less than 2 cm in length) were present over the nets at a mean density similar to that of the larger worms in the control areas. The mudsnail Hydrobia ulvae and the bivalve Macoma balthica were also present at both sites but with higher densities at B3. In neither case were the densities over the mats and control areas significantly different. Corophium volutator, Terebellides stroemi, Ophelia rathkei and Nephtys caeca were also present at B3, but not at B2. For none of these species were the abundances in the experimental and control areas consistently different. 3.2. Blackwater Estuary.
0
control
mulch
net
Fig. 3. The mean density (þs.e., n ¼ 5) of Spartina anglica plants in the control areas and growing through the mats after six months in the Blythe Estuary experiments.
mulchmats from rhizomes that had extended laterally from the adjacent plants. In 2001 site B3 was revisited briefly and although no data could be collected significant growth of S. anglica in the exclusion areas had occurred over the previous
At Maldon the mean depth of sediment accreted on the mulchmats after 5 months was 2.75 cm (0.41 s.e.). These mats were still at the level of the surrounding sediment indicating that little or no deposition had occurred there. In just 5 months these mats were colonised by Spartina anglica plants at a mean density of over 40 m2, a significantly higher number than in the control areas (t ¼ 3.72, d.f. ¼ 8, p ¼ 0.006) (Fig. 6). The control areas had a significantly higher mean density of Nereis diversicolor than over the mats (t ¼ 2.89, d.f. ¼ 8, p ¼ 0.02) (Fig. 6), and these worms were larger
Fig. 4. An example of the development of Spartina anglica marsh over mulchmat (D) and net (E) after 18 months, (and further out, into the area disturbed by treading when samples were collected in June 2000) in comparison to the retreat of the adjacent marsh (A, B, C).
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Site B2
Site B3 3000
no.m-2
3000
2000 1000 0
control
mulch
2000 1000 0
net
1500
6000
4000
4000
1000 500
control
mulch
2000
250 200 150 100 50 0
mulch
control
mulch
net
control
mulch
net
9000
no.m-2
3000 control
mulch
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
control
mulch
net
2000
1000 800 600 400 200 0
250 200 150 100 50 0
6000 3000 0
net
600
600 400 200 0
net
9000
6000
0
mulch
500
0
net
no.m-2
1000 800 600 400 200 0
control
no.m-2
0
control
1000 0
net
no.m-2
0
no.m-2
Terebellides stroemi no.m-2 Ophelia rathkei no.m-2 Nepthys caeca no.m-2
no.m-2
6000
Corophium volutator no.m-2
Hydrobia ulvae no.m-2
2000
1500
2000
Macoma balthica no.m-2
Nereis diversicolor no.m-2
206
control
mulch
net
400 200 0
Fig. 5. The mean densities (þs.e., n ¼ 5) of the invertebrate infauna in the control areas and in the sediment accreted over the experimental mats after six months at the two Blythe Estuary experimental sites.
than those over the mats, as in the Blythe (see above). Hydrobia ulvae were also present at similar densities, of around 600 m2, over the mats and in the control areas. 4. Discussion The infauna exclusion mats at all three locations stimulated significant deposition of sediment and colonisation by Spartina
anglica. Initially, the sediment accretion would have been due to the mats directly enhancing sediment deposition and reducing resuspension, but this effect would have been only temporary because they were rapidly buried. The continued deposition of sediment on the buried mats can be attributed to exclusion of deposit-feeding, mostly by Nereis diversicolor, since no accretion was observed in the adjacent control areas, which remained at the same level as the mats. In the Blythe,
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
Spartina anglica 60
no.m-2
40
20
0
control
mulch
Nereis diversicolor 1200 1000
no.m-2
800 600 400 200 0
control
mulch
Fig. 6. The mean densities (þs.e., n ¼ 5) of the Nereis diversicolor and Spartina anglica in the control areas and in the sediment accreted over the mulchmats after five months at Maldon, Blackwater Estuary.
at Blackshore Mill (the location of B3) French et al. (2000) measured an increase in sediment elevation from September to March but a compensating decrease in the following 6 months. In the Bulcamp Marshes (the location of B2) there was no overall change in the sediment elevations and only 13 cm of sediment had accreted in 55 years (French et al., 2000). In this study the nets at B2 accreted 3 cm in only 6 months. The high rates of sediment accretion in the infauna exclusion areas support the conclusion of Paramor and Hughes (2004, 2005) that N. diversicolor destabilise sediments, by bioturbation and eating the microphytobenthos that help stabilize sediments with their mucus-like secretions, and lower the elevations at which the sediments achieve dynamic stability between the opposing processes of deposition and resuspension. In previous similar exclusion experiments relatively high densities of small Nereis diversicolor and other species were found in the sediment over the mats, perhaps because these areas were a refuge from predation or disturbance by large N. diversicolor (Paramor and Hughes, 2005). A similar result was not found in these experiments perhaps because of their short duration. However, in the control areas at B3, where there was a relatively low density of N. diversicolor, six other invertebrate species were present at relatively high densities,
207
while at B2, where there was a high density of N. diversicolor, only low densities of Hydrobia ulvae and Macoma balthica were recorded. These differences may be interpreted as further evidence that N. diversicolor adversely affects other infaunal species. More directed experiments are required to test this hypothesis. In just a few months the infauna exclusion areas were colonised by Spartina anglica shoots at densities significantly greater than in the control areas, and at B3 even when the adjacent S. anglica bed was dying back. That some S. anglica colonised the control areas at Maldon indicates that S. anglica were developing onto the mudflats, but no such long-term expansion of the edge of the S. anglica marsh has been apparent since these experiments were conducted (RGH, personal observation). The successful colonisation of the exclusion areas by S. anglica, by shoots that had grown upward from rhizomes that had extended underneath the mats from the adjacent plants, indicates that under normal conditions such development of the S. anglica marsh is prevented or restricted by Nereis diversicolor. This may be because the worms disturb or damage developing rhizomes or roots, as was shown with Zostera noltii (Hughes et al., 2000). Emmerson (2000) found that small S. anglica plants transplanted into N. diversicolor exclusion areas were slightly, but significantly, heavier than those transplanted into control areas, and their roots were longer. Spartina anglica, and S. alterniflora in the USA, have suffered from die-back in some areas, leaving sediments and sea walls more open to erosion. The causes of the die-back are uncertain but may include pathogens, grazing or increased waterlogging of the sediment in the centre of Spartina stands. Low oxygen concentrations in waterlogged sediments may reduce meristem growth and lead to increased concentrations of acetylene and sulphide (Gray et al., 1991), or ammonium (Ogburn and Alber, 2006), all of which are toxic to plants. The observation at B3 that the S. anglica marsh was regressing, when the exclusion experiments promoted development of S. anglica, indicates that interactions with infauna may also be a contributory factor to S. anglica dieback, particularly in the south and east of the UK, where Nereis diversicolor has a high abundance high in the intertidal zone (Paramor and Hughes, 2004) and where the loss of S. anglica in the UK has been the greatest. Levin et al. (2006) reported that an invasion of Spartina in San Francisco Bay resulted in a ‘‘trophic shift’’ where deposit feeding polychaetes (including nereids) that consumed Spartina detritus did not decrease in abundance in invaded areas, whereas other invertebrate taxa, including amphipods and bivalves, became less abundant. A similar shift in the infaunal invertebrate community, with increased polychaete dominance, has apparently occurred in SE England, and on the eastern shores of the Southern North Sea, for unknown reasons (Hughes, 1999). The results of Levin et al. leads to the intriguing hypothesis that the increase in abundance of Nereis diversicolor in SE England may be due, in part, to the increased supply of detrital food from the extensive S. anglica plantings (and their
208
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209
subsequent spread) in the first half of the last century. Further, a positive feedback may have been established more recently, where N. diversicolor have additionally benefited from the detritus produced by the Spartina dieback which, this study indicates, may be accelerated by these polychaetes. This is in addition to the loss of other saltmarsh vegetation that they cause (Paramor and Hughes, 2004). A study of the food sources of the high density populations of N. diversicolor in the upper intertidal areas in SE England is required to test this hypothesis. The UK agencies responsible for saltmarsh conservation and restoration, The Environment Agency and English Nature, rely on managed realignment as the main method by which the UK is to try to achieve is international obligations regarding saltmarsh restoration (Morris et al., 2004). The results presented here from the two old realignment areas in the Blythe Estuary extend and support the conclusion of Paramor and Hughes (2005), that infaunal invertebrates will colonise the accreting sediment in low-lying managed realignment sites and will restrict or prevent saltmarsh development. This together with the insufficient availability of land for realignment (Covey and Laffoley, 2002), is why managed realignment is not the means by which the UK will be able to restore sufficient saltmarsh area to meet its obligations under the Biodiversity Action Plan. Morris et al. (2004) disputed this conclusion, but their arguments were flawed and the total area of vegetated saltmarsh created in managed realignment sites since 1995 in SE England is about the same as just 3 years losses (Hughes, in press). Morris et al. (2004) suggested that the lack of vegetation in the Blythe realignment areas after more than 50 years was due to ‘‘the physical profile of the estuary’’ but presented no supporting evidence. The results presented here show this is not correct. These experiments show that managed realignment is not necessary for saltmarsh creation, since the stationary or retreating edges of Spartina anglica marshes can be extended to seaward on existing mudflats. This result also confirms the conclusion of Hughes and Paramor (2004) that the lower limit of saltmarshes in SE England is not determined by sea level, and is, therefore, independent of rising sea level. Consequently the regional-scale decline in saltmarsh cannot be explained by coastal squeeze and managed realignment is not necessary for saltmarsh creation (Hughes and Paramor, 2004). Extending these experiments in scale and geographic range could be a simple and cheap means of managing development of pioneer zone vegetation and sediment accretion, replacing some of the eroding vegetation with economic benefits envisaged by those that originally transplanted S. anglica around this coast. S. anglica is perennial, relatively large, and morphologically more complex than the annual pioneer zone species. Consequently the benefits of increasing S. anglica marsh area, reduced wave impact on existing sea walls and ecological benefits associated with the high primary productivity of S. anglica (Long et al., 1990) would be greater. Deliberately extending S. anglica marshes onto mudflats might seem anathema to those in regions where such growth of
Spartina marshes is an ecological problem (see Section 1) but in SE England saltmarsh conservation would be considered more appropriate, given the continuing high rates of vegetation loss. References Adam, P., 1990. Saltmarsh Ecology. Cambridge University Press, Cambridge UK, 461 pp. Ayres, D.R., Smith, D.L., Zaremba, K., Klohr, S., Strong, D.R., 2004. Spread of exotic cordgasses and hybrids (Spartina sp.) in the tidal marshes of San Francisco Bay California, USA. Biological Invasions 6, 221e231. Burd, F., 1992. Erosion and vegetation change on the saltmarshes of Essex and north Kent between 1973 and 1988. In: Research and Survey in Nature Conservation 42. Nature Conservancy Council, Peterborough. Callaway, J.C., 2005. The challenge of restoring functioning salt marsh ecosystems. Journal of Coastal Research 21, 24e36 (special issue 40). Charman, K., 1990. The current status of Spartina anglica in Britain. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 11e14. Chen, Z.Y., Li, B., Zhong, Y., Chen, J.K., 2004. Local competitive effects of introduced Spartina alterniflora on Scirpus mariqueter at Dongtan of Chongming Island, the Yangtze River estuary and their potential ecological consequences. Hydrobiologia 528, 99e106. Covey, R., Laffoley, D.d’A., 2002. Maritime State of Nature Report for England; Getting Onto an Even Keel. English Nature, Peterborough, UK. Emmerson, M., 2000. Remedial habitat creation: does Nereis diversicolor play a confounding role in the colonisation and establishment of the pioneering saltmarsh plant Spartina anglica? Helgoland Marine Research 54, 110e116. French, C.E., French, J.R., Clifford, N.J., Watson, C.J., 2000. Sedimentationerosion dynamics of abandoned reclamations. Continental Shelf Research 20, 1711e1733. Frid, C.L.J., Chandrasekara, W.U., Davey, P., 1999. The restoration of mud flats invaded by common cord-grass (Spartina anglica, CE Hubbard) using mechanical disturbance and its effects on the macrobenthic fauna. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 47e61. Goss-Custard, J.D., Moser, M.E., 1990. Changes in the numbers of dunlin (Calidris alpina) in British estuaries in relation to changes in the abundance of Spartina. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication No. 2. HMSO, London, UK, pp. 69e71. Gray, A.J., Benham, P.E.M., Raybould, A.F., 1990. Spartina anglica e the evolutionary and ecological background. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A RESEARCH REVIEW. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 5e10. Gray, A.J., Marshall, D.F., Raybould, A.F., 1991. A century of evolution in Spartina anglica. Advances in Ecological Research 21, 1e62. Groenendijk, A.M., 1986. Establishment of a Spartina anglica population on a tidal mudflat: a field experiment. Journal of Environmental Management 22, 1e12. Hughes, R.G., 1999. Saltmarsh erosion and management of saltmarsh restoration; the effects of infaunal invertebrates. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 83e95. Hughes, R.G., 2004. Climate change and loss of saltmarshes; consequences for birds. Ibis 146 (Suppl. 1), 21e28. Hughes, R.G. On the loss of saltmarshes in south-east England and the utility of managed realignment in their restoration. Journal of Applied Ecology, in press. Hughes, R.G., Lloyd, D., Ball, L., Emson, D., 2000. The effects of the polychaete Nereis diversicolor on the distribution and transplanting success of Zostera noltii. Helgoland Marine Research 54, 129e136. Hughes, R.G., Paramor, O.A.L., 2004. On the loss of saltmarshes in south-east England and methods for their restoration. Journal of Applied Ecology 41, 449e463.
O.A.L. Paramor, R.G. Hughes / Estuarine, Coastal and Shelf Science 71 (2007) 202e209 Levin, L.A., Neira, C., Grosholz, E.D., 2006. Invasive cordgrass modifies wetland trophic function. Ecology 87, 419e432. Long, S.P., Dunn, R., Jackson, D., Othman, S.B., Yaakub, M.H., 1990. The primary productivity of Spartina anglica on an East Anglian estuary. In: Gray, A.J., Benham, P.E.M. (Eds.), Spartina anglica e A Research Review. Institute of Terrestrial Ecology Research Publication, No. 2. HMSO, London, UK, pp. 34e38. Morris, R.K.A., Reach, I.S., Duffy, M.J., Collins, T.S., Leafe, R.N., 2004. On the loss of saltmarshes in south-east England and the relationship with Nereis diversicolor. Journal of Applied Ecology 41, 787e791. Ogburn, M.B., Alber, M., 2006. An investigation of saltmarsh dieback in Georgia using field transplants. Estuaries and Coasts 29, 54e62. Paramor, O.A.L., Hughes, R.G., 2004. The effects of bioturbation and herbivory by the polychaete Nereis diversicolor on loss of saltmarsh in south-east England. Journal of Applied Ecology 41, 440e448. Paramor, O.A.L., Hughes, R.G., 2005. The effects of the invertebrate infauna on the early colonisation of managed realignment areas by saltmarsh plants in SE England. Marine Ecology Progress Series 303, 61e71.
209
Percival, S.M., Sutherland, W.J., Evans, P.R., 1998. Intertidal habitat loss and wildfowl numbers: applications of a spatial depletion model. Journal of Applied Ecology 35, 57e63. Smith, D., Hughes, R.G., Cox, E.J., 1996. Predation of epipelic diatoms by the amphipod Corophium volutator and the polychaete Nereis diversicolor. Marine Ecology Progress Series 145, 53e61. Swales, A., Ovenden, R., MacDonald, I.T., Lohrer, A.M., Burt, K.L., 2005. Sediment remobilisation from decomposing cordgrass (Spartina anglica) patches on a wave-exposed intertidal flat. New Zealand Journal of Marine and Freshwater Research 39, 1305e1319. Taylor, C.M., Hastings, A., 2004. Finding optimal control strategies for invasive species: a density-structured model for Spartina alterniflora. Journal of Applied Ecology 41, 1049e1057. United Kingdom Biodiversity Group, 1999. Maritime Species and Habitats, vol. V. Tranche 2 Action Plans. English Nature for the UK Biodiversity Action Group, Peterborough. Wolters, M., Bakker, J.P., Bertness, M.D., Jefferies, R.L., Mo¨ller, I., 2005. Salt-marsh erosion and restoration in south-east England: squeezing the evidence requires realignment. Journal of Applied Ecology 42, 844e851.