Hydrobiologia

, Volume 625, Issue 1, pp 105–116

Disturbance of intertidal soft sediment assemblages caused by swinging boat moorings

  • R. J. H. Herbert
  • T. P. Crowe
  • S. Bray
  • M. Sheader
Primary research paper

DOI: 10.1007/s10750-008-9700-x

Cite this article as:
Herbert, R.J.H., Crowe, T.P., Bray, S. et al. Hydrobiologia (2009) 625: 105. doi:10.1007/s10750-008-9700-x

Abstract

The impact of swinging boat moorings on intertidal benthic assemblages was investigated in a small estuary on the south coast of England. Mooring buoys fixed near low water mark on a muddy shore were attached to 5 m of galvanised steel chain and had not been let for 12 months. Core samples for macro-invertebrates and sediments were taken both within and outside the chain radius of each buoy. The assemblage structure, biomass and abundance of selected bird prey species were examined at a range of scales. The study revealed variation in the impact of mooring buoys relative to control areas at two different times of sampling. Prior to the removal of buoys, the assemblage structure within areas affected by the buoys was found to be significantly different from unaffected areas. The abundance of the amphipod Corophium volutator, an important bird prey species, was significantly less in the areas affected by the buoys. In the second sampling programme (15 months after removal of buoys), the impact of extant buoys remaining in commission was not detectable. Assemblage structure in areas from which buoys had been removed was distinct from control areas which had never had buoys. The removal of mooring buoys clearly affected the assemblage, yet convergence with control areas, indicative of recovery, was not complete after 15 months. It is suggested that the effect of swinging mooring chains scraping over the mud surface may modify sediments favouring the greater prominence of larger particles such as gravel and shell fragments. The ecological impact of swinging moorings on estuarine benthic assemblages in designated protected areas is discussed in the context of other spatial and temporal disturbances.

Keywords

Disturbance Estuaries Boating Intertidal macrofauna Recreation Coastal management 

Introduction

Recreational boating is increasing worldwide (Cicin-Sain et al., 1998; Widmer & Underwood, 2004). The disturbance it may cause to aquatic habitats is perceived to be of conservation concern (Davenport & Davenport, 2006). Research has focussed on issues concerned with marina developments (Turner et al., 1997), water quality (Langston et al., 1994; Matthiessen et al., 1999), disturbance to benthic habitats and sea grass meadows by permanent subtidal moorings and anchoring (Walker et al., 1989; Creed & Amado, 1999; Francour et al., 1999; Backhurst & Cole, 2000; Milazzo et al., 2004), propellers (Uhrin & Holmquist, 2003) and disturbances caused by boat movements (Eriksson et al., 2004). In Europe, much boating activity falls within marine protected areas, yet there have been few ecological studies that have investigated this impact, especially upon intertidal estuarine habitats. Understanding the responses of marine ecosystems to disturbance is a key to predicting their spatial and temporal dynamics (Pickett & White, 1985). The extent of disturbance is known to influence the diversity and composition of benthic assemblages (Connell, 1978; Probert, 1984; Hall et al., 1994; Hall 1994).

The Solent, on the south coast of England, is one of the most popular sailing areas in the world and has seen a growth in moorings of 27% in the past 30 years to currently stand at approximately 24,000 (Solent Forum, 2008). While many vessels are harboured in marinas or deep-water moorings, a large number of boats on swinging moorings ‘dry-out’ at low tide in estuaries. Boats aground on intertidal mud and sand flats occupy potential bird feeding areas, and in some harbours and estuaries this collective footprint may be large. Moreover, moored boats and associated chains may cause scour and mechanical damage to the mud surfaces as they swing around their anchor point, and potentially impact upon the size and composition of invertebrate populations and assemblages that form important bird prey resources.

Although boats could be attached to moorings for considerable time, most swinging moorings will be subject to periods when boats have left the mooring and are out sailing. In the UK, many leisure craft are lifted out for 4–6 months during the winter, and numerous un-let or visitor moorings can normally be found in harbours and estuaries. Additionally, moored navigation marks may cause similar disturbances. The unseen, yet permanent, impact of the mooring is not the boat attached, which may be away for long periods, but the ground tackle and chain that moves over the sea bed in response to changes in wind and tide.

The aims of this investigation were to determine the extent to which swinging moorings impact upon estuarine soft sediment assemblages and to assess recovery following the removal of the moorings. In addressing these aims, we focussed on variation at a range of scales in assemblage structure, total biomass and the abundance of selected species considered important as food for birds.

Materials and methods

Study area

The Medina is a narrow, linear estuary 7.5 km in length with an intertidal area of 66 ha. It is a component of the Solent European Marine Site (European Habitats Directive 92/43/EEC and EU Birds Directive 79/409/EEC) and a Ramsar Site. Tides are semi-diurnal and the mean spring tide range is 3.6 m. The study area known as the ‘Folly’ (Fig. 1) is approximately halfway along the estuary where it is at its widest (0.5 km). Peak ebb currents here are 0.4 m s−1. Surface salinity is 31–34 and there is seasonal dilution to 26 (Withers, 1979). Salinity at the water–sediment interface at High Water is 33–34. The surface water temperature range is between 6 and 20°C.
Fig. 1

Location of study area on the Medina Estuary, Isle of Wight. Samples were taken from two buoys along the ‘inner mooring line’

At the mouth of the Medina is the internationally famous yachting town of Cowes where there are large marinas and deep-water swinging moorings. Pile moorings, whereby boats are tied between posts, and pontoons for local and visiting craft occupy each side of the narrow channel along the lower 4 km of the estuary. In July 2004, 44 boats of between 5.5 and 8.0 m occupied swinging moorings between mean tide level (MTL) and low water spring tide level (LWS). Collectively, these craft, with an average of 6 m of chain attached, are estimated to scour 3% of the mudflat area as they swing with tide and wind. The duration of scour is dependent on mooring height above low tide and the length of chain. It is estimated that over a 12-h tidal cycle during a spring tide, these chains could scrape the mud surface for 7 h. At high tide, relatively little chain would be on the bottom, whereas just prior to the buoy grounding as the tide recedes, most chain could be scraping over the mud.

Other potential disturbances from bait digging and clam collecting were not observed to occur in the immediate vicinity of the study area, although do occur on the estuary.

Fieldwork and sample processing

Two buoys were selected from an ‘inner mooring line’ of 12 small-craft (<6 m length) moorings that dried out between MTL and mean low water neap tide level in the Medina Estuary (Fig. 1). The buoys, 70 m apart, were attached to 2 m of rope (15 mm) and approximately 5 m of galvanised 8 mm steel chain that was attached to concrete blocks buried in the mud. The moorings had not been let during the year immediately prior to the fieldwork but had continued to swing around their anchor point with changes in tide and wind direction.

In September 2000, two separate random patches of five core samples were taken within the chain radius of each buoy and also within a control area 3–4 m beyond the chain radius of each buoy. Patch diameter was 1.5 m; core diameter was 10 cm and sampling depth 15 cm. The distance between patches was at least 3 m. This hierarchical sampling design was chosen so that any difference found between areas with and without buoys could be more confidently attributed to the buoys rather than small-scale variation in assemblage structure (Thrush, 1991; Winer et al., 1991; Underwood, 1997). Samples were processed using a 0.5-mm sieve and the animals preserved in 5% formaldehyde in seawater. Prior to sorting, samples were stained with Rose Bengal and macrofauna were identified to species where possible. Bryozoans were not identified to species level and, being colonial, were not easy to quantify. Where present in a core, usually on the shell of a bivalve, they were given an abundance of 1 for that core. Sub-sampling was carried out for abundant species such as small oligochaete worms.

Following removal of the mooring buoys and ground tackle in July 2001, the sites were revisited in October 2002 utilising the same procedures to determine whether changes consistent with ‘recovery’ (convergence of the benthic assemblages) had taken place. In addition, to ensure that any evidence for recovery was due to decommissioning and not a reflection of temporary changes in assemblage structure, samples were also obtained from two moorings that had remained in commission. Finding that decommissioned sites were similar to unimpacted control sites yet different from sites subject to ongoing impact would provide clearer evidence for recovery (Chapman, 1998).

Estimation of sample biomass

A contractual requirement to preserve specimens necessitated a non-destructive estimation of sample biomass utilising mean dry-weight values for taxa held on a database (Medina Valley Field Centre, Isle of Wight). These measurements were obtained from material dried at 80°C for 48 h.

Sediment analysis

A core of 10 cm diameter was used to obtain samples for particle size analysis from each of the two ‘Buoy sites’ and two ‘Control sites’. Samples were sieved wet over a stack of Wentworth sieves. The finer clay fractions (below 0.063 mm diameter) were not quantified. The organic content of the sediments was determined by placing 10 g sub-samples in a muffle furnace at 450°C for 8 h and measuring loss in mass on ignition. This was the most practical method available to help avoid overestimation of organic content due to loss of structural water in clays (Schulte & Hopkins, 1996; Cambardella et al., 2001).

Statistical analysis of biological data

Differences in assemblage structure

Non-metric mulitdimensional scaling (MDS) was used to produce a graphical representation of the data using the software package PRIMER (Clarke and Warwick, 1994). MDS plots were based on Bray–Curtis similarity measures calculated using square-root transformed data. Permutational multivariate analysis of variance (PERMANOVA) was used to test hypotheses of difference in community structure among groups of samples from different patches, sites and treatments (Anderson, 2001, 2005; McArdle and Anderson, 2001). Two analyses were done: one for data collected prior to removal of moorings and the other for data collected after removal. Prior to removal of moorings, the factors were: Treatment (fixed, 2 levels: buoy versus control); Site (random, 2 levels, nested in Treatment) and Patch (random, 2 levels, nested in the Treatment × Site interaction). After removal of moorings, the factors were the same, but there was an additional level for the factor Treatment (see above).

The PRIMER routine similarity of percentages (SIMPER) was used to identify which species were important in discriminating among samples from the different treatments.

Differences in abundance of individual taxa and total biomass

Variation in sample biomass and abundance of the more common invertebrate species (Tubificoides spp. Cirriformia and Corophium) known to be important prey items for birds (Prater, 1981) was tested separately using hierarchical ANOVA. A separate analysis was done for each variable. Two sets of analyses were done: one for data collected prior to removal of moorings and the other for data collected after removal. The factors involved were the same as those described above for multivariate analyses. There were five replicates. Heterogeneity of variance was tested using Cochran’s test and where necessary, data were transformed.

Results

Prior to removal of moorings

A visual assessment of the mudflats in the vicinity of the moorings showed no obvious evidence of disturbance of the mud surface within the chain radius of each buoy. All samples contained coarser sediments, including small gravel, within a matrix of fine silt and clay. Below the top 10 mm sediments were anoxic. Particles included shells of cockles and other molluscs. The median sediment size class (D50) and interquartile range (IQR) were determined for each sample (Fig. 2). These were larger from samples affected by the buoys (D50 11.2 and 10.1 mm; IQR 21.7 and 19.3 mm) than the unaffected control areas (D50 5.84 and 1.34 mm; IQR 17.64 and 9.76 mm). The mean organic content of samples obtained from buoys was 2.85% (SE ± 0.05) and 2.65% (SE ± 0.35) from control areas.
Fig. 2

Cumulative percentage weight of each sediment size class for samples from within the chain radius of buoys 1 & 2 and the two control sites. Samples sieved over a Wentworth sieve stack; largest diameter (64 mm). The median sediment size class (D50) is indicated by horizontal line. See text for further details

The fauna was typical of that found previously at lower tidal levels within this part of the Medina Estuary (Withers, 1979). A total of 21 taxa were identified in the samples; 19 species occurred in the areas scraped by buoy chains and 15 occurred in the control areas (Table 1a). Epifaunal species attached to stones and shells including the barnacle Elminius modestus and chiton Lepidochitona cinereus were found only in the areas affected by the buoys.
Table 1

Summary of mean densities of species present and the percentage of samples in which they occurred (a) prior to removal of mooring buoys and (b) after removal of mooring buoys

Taxon

Buoys present

Control areas

   

Occurrence (% samples)

Mean density (m−2)

95% CI

Occurrence (% samples)

Mean density (m−2)

95% CI

   

(a)

   

Sagartia troglodytes

20

25.48

22.91

15

31.85

35.66

   

Nemertea

   

(Lineus sp.)

10

12.74

17.18

5

12.74

24.97

   

Anaitides mucosa

0

0

0

30

261.15

288.39

   

Ampherete sp.

5

6.37

12.48

10

12.74

17.18

   

Cirriformia tentaculata

100

3974.57

1260.40

100

2770.73

1065.32

   

Nereis (Neanthes) virens

85

248.41

77.86

90

229.30

58.97

   

Nephtys hombergii

10

12.74

17.18

5

6.37

12.48

   

Tubificoides benedii

100

5687.96

1233.79

95

3681.57

1252.21

   

Tubificoides sp.

100

8019.20

4062.49

100

7445.95

3798.74

   

Carcinus maenas

5

6.37

12.48

5

6.37

12.48

   

Corophium volutator

100

2751.62

660.95

100

4624.26

675.36

   

Cyathura carinata

5

6.37

12.48

0

0

0

   

Elminius modestus

25

783.45

1444.28

0

0

0

   

Gammarus sp.

5

6.37

12.48

0

0

0

   

Melita palmata

5

6.37

12.48

0

0

0

   

Cerastoderma edule

40

63.70

38.43

25

31.85

24.80

   

Hydrobia ulvae

0

0

0

10

38.22

54.64

   

Lepidochitona cinereus

15

19.11

20.45

0

0

0

   

Littorina littorea

5

6.37

12.48

0

0

0

   

Macoma balthica

55

76.43

33.40

20

25.48

22.91

   

Bryozoa indet.

10

12.74

17.18

0

0

0

   

Taxon

Buoys present

Control areas

Buoys removed

Occurence (% samples)

Mean density (m−2)

95% CI

Occurence (% samples)

Mean density (m−2)

95% CI

Occurence (% samples)

Mean density (m−2)

95% CI

(b)

Cereus pedunculatus

0

0

0

0

0

0

10

35.37

47.71

Nematoda

15

70.74

81.09

20

61.89

57.75

50

335.99

221.93

Nemertea

0

0

0

5

17.68

34.66

15

53.05

56.79

Ampharete acutifrons

0

0

0

5

17.68

34.66

40

159.15

93.75

Anaitides mucosa

5

17.68

34.66

0

0

0

10

53.05

75.85

Tubificoides benedi

80

1167.14

605.79

80

592.41

258.92

95

3713.62

1458.86

T. pseudogaster

90

5110.64

2367.12

85

2749.84

1343.86

95

3890.45

3824.07

Cirriformia tentaculata

100

15066.67

3981.67

100

14169.21

3846.86

100

12767.76

2770.05

Caulleriella sp.

100

2581.85

609.95

70

618.94

300.43

100

1768.39

548.61

Nereis (Neanthes) virens

10

35.37

47.71

0

0

0

5

17.68

34.66

Mediomastus fragilis

5

17.68

34.66

0

0

0

5

17.68

34.66

Cossura longocirrata

5

17.68

34.66

5

17.68

34.66

0

0

0

Capitella capitata

5

17.68

34.66

10

22.10

35.28

0

0

0

Streblospio shrubsolli

35

141.47

92.73

65

579.15

251.29

0

0

0

Carcinus maenas

10

35.37

47.71

0

0

0

35

123.79

75.85

Melita palmata

0

0

0

0

0

0

25

212.21

221.51

Ampharete baltica

0

0

0

5

35.37

69.32

0

0

0

Magelona mirabilis

5

17.68

34.66

0

0

0

0

0

0

Melinna palmata

10

35.37

47.71

5

17.68

34.66

0

0

0

Cerastoderma edule

35

371.36

268.36

80

477.46

161.69

90

1131.77

364.74

Macoma balthica

10

35.37

47.71

20

84.00

82.90

5

17.68

34.66

Abra tenuis

10

35.37

47.71

0

0

0

5

17.68

34.66

Lepidochitona cinerea

0

0

0

0

0

0

10

35.37

47.71

Hydrobia ulvae

90

1697.65

536.01

95

2055.75

536.50

95

2723.32

1030.77

Crepidula fornicata

0

0

0

0

0

0

5

17.68

34.66

Data based on a total of 20 samples taken from patches and sites in each treatment. Densities converted to m2

CI confidence interval

Prior to removal of buoys, samples from different patches within a site and different sites within a treatment (i.e. buoy versus control) were intermingled, indicating no strong spatial patterns of community structure at either of those scales (Fig. 3, Table 3a). There was, however, a significant separation (< 0.024) between samples from areas with buoy chains and samples from control areas (Fig. 3, Table 2a). It should be noted, however, that the stress value associated with this 2D representation is >0.2. Care should therefore be taken interpreting the figure (Clarke, 1993).
Fig. 3

MDS ordination of assemblages in all samples collected prior to removal of moorings. Each point represents a single sample. White triangles and inverted triangles represent samples from buoys 1 and 2, respectively. Grey circles and diamonds represent samples from control sites 1 and 2, respectively. Symbols with and without black centres distinguish samples from different patches within each site. See text for further details

Table 2

PERMANOVA analyses of differences between patches, sites and treatments: (a) prior to removal of moorings (Treatments = buoys versus controls) and (b) after removal of moorings (Treatments = extant buoys versus controls versus buoy removal areas)

Source of variation

df

MS

Pseudo-F

P(MC)

(a)

Treatments = Tr

1

1622.6

4.04

0.024

Sites = Si(Tr)

2

401.7

0.88

0.554

Patches = Pa(Si(Tr))

4

454.2

1.09

0.378

Residual

32

415.2

  

Source of variation

df

MS

Pseudo-F

P

(b)

Treatments = Tr

2

5194.9

2.47

0.033

Sites = Si(Tr)

3

3172.5

1.03

0.469

Patches = Pa(Si(Tr))

7

7211.7

2.38

0.001

Residual

47

20,333

  

Data were square-root transformed. Analyses were done on Bray–Curtis similarity matrices using 999 permutations of residuals under a reduced model. In analysis (a), Monte Carlo tests (MC) were used given the limited number of unique permutations for factor 1

SIMPER analysis revealed 90% of variation among groups of samples collected near buoys and from control areas were caused by differences in abundance of Tubificoides sp., Tubificoides benedii, Cirriformia tentaculata, Elminius modestus, Neanthes virens, Macoma balthica and Cerastoderma edule which occurred in greater abundance among buoys than in control areas and by Corophium volutator and Anaitides mucosa which occurred in greater abundance in control areas than among buoys.

There were no significant differences between patches, sites or treatments for total biomass or densities of Cirriformia tentaculata or Tubificoides spp. (Fig. 4a, b, d; Table 3a, b, d). However, Corophium volutator was significantly more abundant at control sites than at sites with buoys (Fig. 4c, Table 3c; Student–Newman–Kuels (SNK) procedure, < 0.05). The mean density of Corophium was reduced by 40% from 4,624 m−2 in control sites to 2,752 m−2 in sites with buoys.
Fig. 4

a Mean total biomass of macrofauna per core, b density of Cirriformia tentaculata, c density of Corophium volutator and d density of Tubificoides spp. at the sites sampled prior to the removal of mooring buoys. In each graph, each site is represented by two bars. Each bar represents a single patch at which five replicate cores were taken. Mean + SE shown. Note different scales on the y axes

Table 3

Analyses of variance of (a) total biomass, (b) density of Cirriformia, (c) density of Corophium volutator and (d) density of Tubificoides spp.

Source

df

(a)

(b)

(c)

(d)

MS

F

MS

F

MS

F

MS

F

Treatment = Tr

1

0.07

0.62

2.35

2.26

35067544

58.40*

66545812

0.42

Site = Si(Tr)

2

0.12

1.46**

1.04

0.98

600443

0.16

1.57 × 108

1.78***

Patch = Pa(Si(Tr))a

4

0.08

0.97

1.06

1.60

3790099

1.69

96140391

1.09

Residualb

32

0.08

 

0.66

 

2249027

 

87498057

 

Prior to removal of mooring buoys

Variance heterogeneity was tested using Cochran’s test. Variances were homogeneous except in analysis (b). Data for that analysis were transformed, X′ = log (X + 1). After transformation, Cochran’s C = 0.25, n.s.

* Significance at < 0.05; ** tested over pooled MS (a + b = 0.08 with 36 df); *** tested over pooled MS (a + b = 88458316 with 36 df)

After removal of moorings

In October 2002, 15 months after removal of selected moorings, 15 species occurred in the control sites, 19 in sites from which buoys had been removed (decommissioned) and 18 species occurred in areas that were still being scraped by buoy chains (Table 1b). Generally, percentage occurrences and average densities of most species were comparable at sites in the three treatments. There were however differences in the densities of single species. The burrowing anemone Cereus pedunculatus was found only in decommissioned sites where it attained a mean density of 35 m−2 (Table 1b). Although cockles, Cerastoderma edule, were on average two to three times more abundant at decommissioned sites compared to controls or extant moorings (Table 1b), significant variation was at the scales of Patches and Sites, rather than among treatments (Treatment: F2,3 = 1.33, > 0.38). Of the common species considered important as food for birds, there were no differences among treatments in the abundances of Tubificoides spp. (Treatment: F2,3 = 2.05, > 0.47) or Cirriformia tentaculata (Treatment: F2,3 = 0.14, > 0.87). The burrowing amphipod Corophium volutator, which was less abundant in areas affected by the original moorings in September 2000, was not recorded at any site during this sampling.

After removal of moorings, there was considerable variation in assemblage structure at the scale of Patches and among Treatments (Fig. 5, Table 2b), although the stress is again >0.2, so the specific placement of the points should be interpreted with caution (Clarke, 1993). Assemblages in areas from which moorings had been removed were distinct from those in control areas in which buoys had never been present (PERMANOVA post hoc pairwise comparisons). SIMPER analysis indicated that this difference was underpinned by reduced abundances of Cirriformia tentaculata in areas from which moorings had been removed and increased abundances of Tubificoides benedi, T. pseudogaster, Caulleriella sp. and Hydrobia ulvae. Assemblages in control areas and areas with buoys still in place were not distinguishable from each other.
Fig. 5

MDS ordination of assemblages in all samples collected after removal of moorings. Each point represents a single sample. Pale grey triangles and inverted triangles represent samples from extant buoys (P1 and P2); dark grey squares and diamonds represent samples from control sites (C1 and C2); black circles and open circles represent samples from sites from which buoys had been removed (R1 and R2)

Biomass of macrofauna varied significantly from patch to patch and from treatment to treatment (Patch: F6,48 = 5.59, < 0.001; Treatment: F2,3 = 19.58, < 0.05). Mean biomass at decommissioned sites was significantly greater than at control sites and sites at which buoys were still present (78.5 g m−2 v 13.3 g m−2 v 14.6 g m−2 respectively; SNK procedure, < 0.05). In control areas, densities of several species, notably Cerastoderma edule, Hydrobia ulvae, Cirriformia tentaculata and Tubificoides spp., were considerably different to the initial survey in September 2000 prior to removal of buoys.

Discussion

This study has revealed variation in the impact of mooring buoys relative to control areas at two different times of sampling. In the first sampling programme (prior to removal of buoys), there were clear differences between areas with and without mooring buoys. In the second sampling programme (15 months after removal of buoys), the impact of the buoys was not detectable. After the removal of buoys, the total biomass in areas from which buoys had been removed was far greater than in control areas which had never had mooring buoys and from extant mooring buoys. Assemblage structure in areas from which buoys had been removed also diverged from that in other areas and was statistically distinct from control areas which had never had buoys. The removal of mooring buoys has clearly affected the assemblage. At the time of sampling, however, the assemblage had not converged with control areas, suggesting that if recovery is underway it was not complete after 15 months.

For individual species and assemblage structure, variability was evident at a range of spatial scales, from individual cores separated by tens of centimetre, through patches separated by metres to sites separated by tens of metre. Such variation is common in sedimentary habitats (Morrisey et al., 1992; Hall et al., 1994; Kendall & Widdicombe, 1999). In some cases variation was related to the presence, absence or removal of buoys and in others it was not. Mooring buoys clearly have some impact on the macrofauna of the Medina Estuary, but there are other sources of spatial and temporal variation which sometimes have a greater impact (Summers, 1980; Thistle, 1981; Savidge & Taghon, 1988; Cadée, 1990; Raffaelli et al., 1990). Moreover, deposition and erosion of sediment are likely to vary over at least annual time scales. Although the number of sediment samples was limited, it is suggested that the effect of the swinging mooring chains scraping over the mud surface may modify sediment composition favouring the greater prominence of larger particles such as gravel and shell fragments. These were certainly more evident in the sediment samples obtained from within the chain radius of the buoys.

Some larger polychaete and bivalve species may have been undersampled with a 10-cm diameter corer. Because of this potential size-bias, densities of some species may have been calculated as significantly higher or smaller than those commonly found in UK estuaries. The abundance of many species, e.g. Tubificoides spp., Nereis (Neanthes) virens, Cirriformia tentaculata, which are likely to be important prey items for wading birds (Prater, 1981), tends to be greater amongst the buoys. However, the tube dwelling amphipod Corophium volutator, a filter and deposit feeder on the upper 2 cm of mud surface (Meadows & Reid, 1966; Mermillod-Blondin et al., 2005), was significantly less abundant amongst the buoys compared to control areas. It is possible that frequent scraping by chains could damage burrows or modify sediments preventing adequate construction. Corophium was not recorded at any sampling site after removal of buoys; vagaries of life cycle are probably responsible for what it is likely to be only a temporary absence of this generally common species, although interactions with other species and/or changes in background sediment composition are also possible (Hughes & Gerdol, 1997; McCurdy et al., 2005). Differences in the abundance of particular species may be due to changes in the chemical and physical properties of the mud, such as the degree of anoxia and drainage, caused by particle size variability. They could also be due to competition between and within species or to differential predatory activity by birds and fish (Cadée, 1990; Raffaelli et al., 1990). For example, populations of some species may be higher due to reduced predation: fish may be deterred by the movement of chains and some birds may avoid the brightly coloured mooring buoys. Within the chain radius of the buoy, there may be temporal variability in the extent of disturbance and rate of recovery due to the interaction of wind direction, tidal movement and use of the mooring. More complex interactions may be occurring whereby localised small-scale disturbances on mudflats, caused by foraging by predators within areas of high prey density, accentuate the degree of patchiness (Hall et al., 1994). However, in the initial analyses, small-scale patchiness is approximately similar in the vicinity of both buoys and controls: there were no significant differences in the abundance of particular species between sites or patches within either of the treatments. Prior to the cessation of use of these moorings, disturbances of the mud surface may have affected benthic assemblages within control areas outside the chain radius of the buoys and it is possible that these areas may still be recovering.

Although convergence between areas from which buoys had been removed and control areas was not apparent 15 months after decommissioning, there was evidence that assemblages are changing in areas from which buoys had been removed. It is not clear whether convergence will occur or within what time frame. In a study of the recovery of soft sediment assemblages, following physical disturbances of different intensity (Dernie et al., 2003), the fauna within experimental plots took between 64 and 208 days to converge to that of surrounding control areas. The Medina appears to be on a slower trajectory. Given the high level of temporal variation in the system, indicated by the changes observed in the control areas over the two year period, it would be necessary to collect data on a number of occasions prior to and after removal of buoys over an extended period to generate clear-cut evidence of recovery (Chapman, 1998).

Many of the invertebrate species found in this study are important prey items for wading birds (Prater, 1981). The amphipod shrimp Corophium volutator and oligochaete worms (Tubificoides spp.) are especially favoured by smaller waders such as Redshank and Dunlin. Polychaetes, such as Cirriformia tentaculata, Nereis (Neanthes) virens and Nepthys hombergii, and molluscs, Cerastoderma edule and Macoma balthica, are also regarded as essential prey items for larger species such as Oystercatchers, Curlew and Godwits (Burton, 1974; Prater, 1981). A 40% reduction in abundance of Corophium in the vicinity of moorings reduces the potential food resource for various species within the marine protected area. Personal observations suggest that foraging does occur in close proximity; Turnstone (Arenaria interpres) was observed feeding both within and beyond areas affected by mooring buoys. In September, when the samples were obtained, there were intermittent large flocks of wading birds on passage in the vicinity. Prater, (1981) suggests depletion of invertebrate stocks occurring from July onwards; however, it would be surprising if significant reduction in prey density within these areas had occurred so early when bird numbers were still relatively low.

The scope of this study was limited to the immediate vicinity of the buoys. The overall ecological impact of chain-scouring on the quality of designated habitat is difficult to quantify without more detailed sediment maps, and there could be interactions with a variety of other disturbances. Even if the locally exaggerated disturbances caused by chain-scouring result in habitat modification, these habitat types may be commonplace in undisturbed parts of the Medina Estuary and elsewhere in the marine protected area. If this is the context, then the impact of the buoys may be considered to be negligible. However, human-induced disturbances of the kind examined may not be acceptable in terms of maintaining favourable habitat or for the protection of scarce species. For example, while these habitats may encourage some birds, they may not be attractive to Black-tailed godwits (Limosa limosa), for which the Solent and Medina Estuary have been specially designated, that require a variety of food items including Corophium (West et al., 2007).

Scouring caused by anchor chains is just one of several possible impacts of a swinging mooring. The disturbance impact caused by movement of the hull and keel of tethered boats has not been examined and will vary considerably depending on vessel size, hull shape and keel type. The type of impact will also depend on substrate and tidal regime. With increasing pressure for space within designated conservation areas, the impact of different boat mooring configurations may need to be examined and mitigation approaches considered. On the Medina, six intertidal swinging moorings from the inner mooring line and three from the outer mooring line (Fig. 1) were re-laid below extreme low water spring tide mark to offset reclamation of mudflat and dredging disturbances in the upper estuary. In areas where swinging moorings are scattered throughout the intertidal region, zoning schemes that concentrate moorings within defined areas would create larger areas of undisturbed mudflats. Holding boats in line between fixed buoys or ‘trot’ type moorings would significantly reduce scour effects where this might be considered a problem.

Acknowledgements

The work was supported by the Environment Agency via the Isle of Wight Estuaries Project. Thanks to Keith Marston, Steve Thompson and Graeme Leggatt for field and laboratory assistance. We are grateful for the constructive comments and suggestions of referees.

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  • R. J. H. Herbert
    • 1
    • 2
  • T. P. Crowe
    • 3
  • S. Bray
    • 4
  • M. Sheader
    • 5
  1. 1.Medina Valley CentreNewportUK
  2. 2.School of Conservation SciencesBournemouth UniversityPooleUK
  3. 3.School of Biology and Environmental Science, UCD Science Centre WestUniversity College DublinDublin 4Ireland
  4. 4.School of Civil Engineering and the EnvironmentUniversity of SouthamptonSouthamptonUK
  5. 5.National Oceanography CentreSchool of Ocean & Earth Sciences, University of SouthamptonSouthamptonUK

Personalised recommendations