ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on January 19, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(2):309-317; doi:10.1093/icesjms/fsl037
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Can traditional harvesting methods for cockles be accommodated in a Special Area of Conservation?
1 Quercus, Queens University of Belfast, School of Biological Sciences, 97 Lisburn Road, Belfast BT9 6BL, UK
2 Queens University of Belfast, School of Biological Sciences, 97 Lisburn Road, Belfast BT9 6BL, UK
Correspondence to M. P. Johnson: tel: +28 90972297; fax: +28 90975877; e-mail: m.johnson{at}qub.ac.uk
McLaughlin, E., Portig, A., and Johnson, M. P. 2007. Can traditional harvesting methods for cockles be accommodated in a Special Area of Conservation? – ICES Journal of Marine Science, 64: 309–317.The European Natura 2000 project attempts to balance conservation and exploitation by permitting activities that do not affect the conservation status of designated sites. Given the scale of Natura 2000, guidelines are needed to facilitate the drafting of simple site management plans. This need is particularly acute for traditional harvesting methods for which there is usually strong local opposition to the imposition of controls. These issues were examined in Strangford Lough, a special area of conservation where cockles have traditionally been harvested by hand-raking. Raking was found not to affect the ability of cockles to rebury. There were significant reductions in Zostera biomass when raking was carried out within eelgrass beds (a 90% reduction in biomass available to winter migrant birds from summer raking). Traditional harvesting methods could therefore be accepted in Strangford as long as Zostera beds are avoided. A relatively low intensity of harvesting activity in Strangford Lough probably reflects low cockle densities (average 91.8 m–2), with the most economically valuable individuals at some distance from points of access to the shore. An economically feasible management plan could sanction traditional harvesting and result in the implementation of more resource-intensive management only if increases in cockle stocks and market prices stimulate large increases in harvesting activity.
Keywords: burial, conservation, disturbance, habitats directive, shellfish, unregulated harvesting, Zostera
Received 9 May 2006; accepted 24 November 2006; advance access publication 19 January 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
The coastal environment is subject to a multiplicity of uses that can interact to cause large changes in ecosystem structure and the sustainability of resource use (Lotze and Milewski, 2004). A widely applied management tool in the face of multiple pressures is to create reserves or marine protected areas (MPAs). More than 5000 MPAs have been designated to date (Wood, 2005). The EU Habitats Directive (European Council Directive 92/43/EEC, 1992) is an example of this approach. This legislation aims to create a network of protected areas (so-called Natura 2000) throughout Europe for the purposes of biodiversity conservation. Individual sites in the Natura network are designated as Special Areas of Conservation (SACs). Designation as an SAC does not, however, automatically establish a no-take zone from where fishing activity is banned. Exploitation of an SAC's resources is permitted under the Habitats Directive as long as the favourable conservation status of the site is not threatened. The coexistence of exploitation and conservation, for example, using catch-and-release schemes rather than no-take management, offers opportunities for conservation bodies to overcome the opposition of many groups to the imposition of MPAs (Salz and Loomis, 2004). Development of such approaches into co-management schemes, in which fishers have some ownership of the exploited resource, are thought to be effective in sustainable fisheries (Castilla and Defeo, 2001). Clearly, if fishers and conservationists are to have confidence in management plans, scientific estimates of potential impacts are of key importance. This study is concerned with the potential effects of traditional methods of cockle harvesting (hand-raking) in Strangford Lough, an SAC primarily designated for a number of habitats, including intertidal mudflats. The need for assessments of impacts in European SACs and other reserves is acute. More than 180 sites designated as SACs contain mud- and sandflats in the European Union's proposed list for the Atlantic region (European Council, 2004). This is a number beyond the likely capacity for regulatory agencies to run detailed site-specific management schemes.
In the absence of informed consensus, conservationists and fishers can become disillusioned with protected areas. This process may be particularly emphasized for SACs where exploitation of some natural resources may be permitted if it does not alter the favourable conservation status of the site. Conservationists may object to the continuation of exploitative activities under the precautionary principle (that the avoidance of undesirable consequences, even if the risk of these is unknown, should take precedence in the use of the area). Alternatively, there may be a desire to return the protected area to a state that existed before exploitation. In contrast, fishers may perceive regulations associated with protected areas as an unnecessary restriction on personal liberty. These attitudes are clear in the evidence presented to the committee considering cockle-harvesting legislation for Northern Ireland (Anon., 2000). Submissions to the committee included details of requests that unregulated shellfish gathering be stopped, as well as requests that the legislation does not interfere with traditional practice.
When cockles (Cerastoderma edule) are raked by hand from the sediment, there are potential impacts on the sediment community, on cockles, as a result of direct effects through trampling and through indirect effects such as disturbance to foraging birds. The effects of hand-raking on benthic communities appear relatively short-lived (Kaiser et al., 2001; MacKenzie and Pikanowski, 2004; Mistri et al., 2004). An exception to this pattern may occur for longer lived species such as eelgrass, Zostera (Kaiser et al., 2001). Spatial models have indicated that bird survival is not reduced in the presence of adequately controlled cockle harvesting (Stillman et al., 2001). Estimates of hand-rake harvesting impacts on cockle populations are less well known (Gubbay and Knapman, 1999). Owing to the unregulated nature of the fishery in many areas, there are often no reliable catch statistics (Roberts et al., 2004). This, combined with unpredictable recruitment and mortality, makes estimates of a fishery's impact difficult in all but the most extreme cases (Piersma et al., 2001). We therefore sought to address three questions: (i) does the spatial pattern of cockle densities have implications for the potential impacts of traditional harvesting?; (ii) does the physical disturbance of raking affect the ability of cockles to avoid predation by reburrowing?; (iii) does raking affect intertidal Zostera beds?
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Strangford Lough is a National Marine Reserve, with the most recent designation being that of an SAC, through the Habitats Directive (European Council, 1992). One of the important habitats that the Lough is designated for is mudflats and sandflats not covered by seawater at low tide, for which Strangford is considered to be one of the best areas in the United Kingdom. Zostera and cockle beds are specifically mentioned as components of these intertidal mudflats in the relevant SAC documentation (http://www.jncc.gov.uk/protectedsites/sacselection/sac.asp?EUCode=UK0016618). Traditional hand-rake harvesting has been observed in most of the mud- and sandflat areas in Strangford. To assess spatial patterns in the cockle population within this area of mudflats and sandflats (Figure 1), a broad-scale survey was carried out during July and August 2004. Material was collected using a sampling grid with a separation of 200 m between points, leading to >400 samples throughout the Lough. Samples were taken from the north end of the Lough because this is the area with the most extensive mudflats and cockle beds. A 0.0625 m2 quadrat was placed at each sample point (located using a portable Global Positioning System). Sediment in the quadrat was removed by hand to a depth of 5 cm. This material was sieved (2 mm mesh), and all cockles were removed and placed in a labelled polythene bag. The length of each collected cockle was measured (as the longest distance between the anterior and posterior shell margins; Jensen, 1992) using vernier callipers (0.1 mm precision). Shell ages were estimated by counting the number of winter growth rings on the shell's surface (Richardson et al., 1980). Ages beyond 12 annual rings were not estimated because of problems in distinguishing bands in older shells. The final age class therefore represents shells with at least 12 winter bands.
|
The spatial structure of age classes in cockle distribution was assessed using a multivariate technique. The main aim of this analysis is to define the spatial scale over which patches of cockles have a coherent age structure. The differences between all possible pairs of quadrats were summarized using normalized Euclidian distances. The Euclidian distance between points in two dimensions can be found using Pythagoras' theorem for a triangle. It is straightforward to extend this to Euclidian distances across the 13 dimensions associated with separate age classes. Counts in each age class were normalized (mean zero, s.d. 1) to give all age classes equal weighting. A normalized Euclidian distance of zero implies that quadrats have an identical age structure. Large normalized Euclidian distances imply greater differences between the age-class structures of quadrats being compared. The spatial structure in age-class distributions is summarized by calculating average Euclidian distances between quadrats over a range of separations between quadrats. If there is no spatial structure in the cockle bed, the average Euclidian distance will not change with separation distance. Summary plots of Euclidian distances can be interpreted in the same manner as variograms. A general pattern is for variograms to have non-zero variance at the lowest separation between locations (the nugget) and for estimates of variance to increase until a background level is reached (Legendre and Legendre, 1998). The range over which estimates of variance are relatively low defines the scale of autocorrelation (spatial structure) in a variogram. Hence, relatively low Euclidian distances at small separations suggest that locally coherent cockle age structures exist across the range of sites where the relatively low mean values are found.
The effect of hand-raking on cockle reburial was assessed in a field experiment in the northeast of the Lough. Sediments there are typical of the sheltered shores of the Lough: a high proportion of silt and other fine particles. Two factors were investigated in an orthogonal design: the disturbance caused by raking individuals and the disturbance caused by raking the sediment. Both these factors have the potential to increase reburial times. There were two levels of manipulation of individual cockles: raked and control. Raked cockles were subjected to hand-raking, removed from the sediment, and sieved, as would be the practice in traditional harvesting. Control cockles were gently removed from the sediment by hand, but received no further handling before being allocated to experimental treatments. Sediment manipulations were also divided into a rake treatment and a control. Before the start of the experiment, 0.25 m2 areas were outlined. Rake treatments were then applied to half the experimental plots, following random allocation of treatments. Raked plots mimicked small-scale harvesting, with surface raking of sediment (to 5 cm depth) within the 0.25 m2 area and removal of resident cockles. In the control areas the sediment was undisturbed. The two levels of each factor result in a total of four treatments: (i) raked cockles and raked sediment; (ii) raked cockles and control sediment; (iii) control cockles and raked sediment; (iv) control cockles and control sediment.
Treatments were replicated four times. Estimates of reburial were made by haphazardly scattering 25 cockles across each plot. The number of buried cockles was recorded after 1 h. All experimental cockles were then removed to measure shell heights. As this was an impact study, estimates of statistical power (Zar, 1996) were made to determine the likely size of the impact that would have been detected given the level of replication used.
The physical disturbance of hand-raking on Zostera beds was investigated using field experiments at two sites on Strangford Lough, Ringneil Bay, where Zostera noltii was dominant, and the North End, which has a mixed bed of Z. noltii and Zostera angustifolia. In all, 12 sites (4 m2) were marked out at each site with canes and randomly allocated to treatments. Harvest treatments (n = 6) were raked over to remove cockles. Control treatments were left untouched. The initial manipulation was carried out in June 2004, and data were collected on Zostera biomass 3 months later, shortly before the main arrival of winter migrants such as Brent geese. Biomass estimates were made on the basis of a randomly located core sample for each replicate (22.5 cm diameter, 398 cm2 area) taken to a depth of 5 cm. Collected material was sieved using a 2 mm metal mesh, and all Zostera biomass was placed in a labelled polythene bag, frozen, and stored. The samples were defrosted in warm water, sieved to remove adhering debris and sediment, blotted dry, wet-weighed, dried for 48 h at 60°C, and then dry-weighed. Two-way analysis of variance (ANOVA) was used to determine the effect of hand-raking on Zostera wet and dry biomass.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Cockles were found throughout the survey area, and although some 30% of quadrats contained no cockles, such areas without cockles were relatively small (Figure 2). The average density of cockles in samples was 91.8 m–2 (s.e. 6.99), with a maximum density recorded of 944 m–2. The higher densities (>100 m–2) of older cockles tended to be found towards the low water mark, along the west side of the Lough. Although high densities of recruits and older cockles were only found together at two sites (Figure 3), there was no overall trend for recruitment to be lower in the presence of high adult densities. Correlations between recruits and older cockles were not significant and close to zero (Table 1). Pooling older cockle densities (all individuals aged 4+) still provided no evidence for segregation of recruits and adults (rs = –0.008, p > 0.5). The clearest pattern in the age-class distribution was for the densities of cockles with a similar band number to be correlated across separate cores.
|
|
|
There was a relatively high variance in the age structure of cockle samples among adjacent quadrats. The average normalized Euclidian distance among quadrats 200 m apart was 85% of the mean Euclidian distance at all scales up to 3 km. Despite this small-scale variability, there was still some spatial structure in the cockle beds at scales up to
400 m (Figure 4). These results suggest that a central quadrat and its eight nearest neighbours could be conservatively taken as a coherent "patch" for estimates of population dynamics, including age-class distributions. Summaries at this scale suggest that a number of different age-class patterns exist in the Lough (Figure 5). The overall age structure suggests some variability in recruitment, but a gradual decline in abundance with increasing age over 2 years. This contrasts with smaller scale patterns that include patches dominated by recruits (Figure 5b) and areas with distinct age classes (examples in Figure 5c and d). Modal age classes differed between patches. For example, the separate patches shown in Figure 5 contain either a relative peak in abundances at ages 3 and 4 or a trough.
|
|
Shore heights relative to mean high-water springs (MHWS) were available for the northwest region of the Lough. The average age of cockles (per quadrat) there increased significantly at lower levels on the shore (Figure 6a). The fitted curve was a better descriptor of the change in average age than a straight line, even after adjusting for the additional degrees of freedom in a quadratic polynomial (both regressions significant at p < 0.05; r2adj = 62.5% for the linear fit, r2adj = 68% for the polynomial). There appeared to be more rapid growth of cockles low on the shore. Although recruits were not significantly larger at increasing depths below MHWS, cockle lengths within given age bands (ages 1–6) increased significantly with decreasing level on the shore (Figure 6b). The r2 values for significant regressions were 32.6% for year 1, 26.7% for year 2, 34.1% for year 3, 49.2% for year 4, 25.4% for year 5, and 10.6% for year 6. Cockles >6 years old were infrequent in quadrats where the height was known (n = 32; no cockles >8 years old). For cockles aged 7 and 8, there was no consistent pattern of shell length with shore height. This may be because of the small sample size and the absence of older cockles in shallow water (<3.7 m).
|
Hand-raking did not significantly affect the cockles' ability to rebury (Table 2, Figure 7). A power analysis indicated that a 26.1% difference in mean burial rates would have been detected in 80% of experiments with four replicates. The data were also analysed with respect to small cockles (<20 mm) because these are more likely to be discarded by harvesters and left on the surface of the sediment. Again, there were no significant differences between treatments, according to ANOVA (Table 2). Small individuals had a faster mean rate of burial than larger cockles (51.7% for small cockles, 31.1% for all cockles).
|
|
Raking the sediment led to a significant reduction in Zostera biomass after 3 months (Table 3, Figure 8). On average, there was more than three times the biomass in unraked plots than in raked treatments. The site effect reflects the greater biomass at Ringneil on both raked and unraked plots.
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Hand-raking appears to be a relatively low-impact activity that may be accommodated in conservation areas as long as Zostera beds are avoided. The Burry Inlet in Wales is thought to be sustainably managed by limiting the number of harvesters and restricting the techniques used to collect by hand (Stillman et al., 2001; West et al., 2003). Such a management scheme could be applied to Strangford, involving restrictions on the number of individuals permitted to harvest. Precise estimates of the numbers of people involved in cockle harvesting in Strangford are not available (Roberts et al., 2004), although the maximum number observed on any day is <20 (AP, pers. obs.). The relatively few harvesters (compared with 50 licences in the Burry Inlet and up to 2000 harvesters in the Three Rivers fishery in Wales; Nautilus, 2000) may indicate that cockle harvesting in Strangford is a marginally profitable activity. The average cockle density is less than would cause fishery closures in other areas. For example, the management plan for the Dee Estuary in North Wales (Environment Agency, 2001) closes the cockle fishery when average densities of 25 mm cockles are <100 m–2.
Alongside the relatively low densities of cockles, their spatial distribution may also reduce the motivation for people to engage in traditional harvesting. The most accessible area of shore is in the southeast of the sample area, close to Greyabbey, where cockles >1 year old were at a very low density (see the age structure in Figure 5b). Although this pattern could reflect overfishing, an average mortality of 97% was estimated between May and August 2004 in marked plots on the east side of the Lough (with many gaping or empty shells on the surface, indicating that mortality was not attributable to harvesting activity). Such high mortalities would lead to a recruit-dominated age structure. The largest densities of harvestable cockles are generally some distance from access points to the shore. Few harvesters will choose to walk a potential round-trip distance of 4 km to access these beds in the northwest of the Lough (mechanized transport is not permitted under the Northern Ireland Fisheries Amendment Bill of 2001). In the northwest of the mudflats, cockle distributions reflect patterns seen elsewhere: greater densities of large cockles towards the lower shore, with more rapid growth under conditions of increased tidal immersion (Sanchez-Salazar et al., 1987; Jensen, 1992; de Montaudouin, 1996). Recruitment preferences, post-recruit movement, and differential mortality could all contribute to these patterns. There was little evidence that recruits avoided higher densities of adults. This may reflect the relatively low densities in the Lough compared with densities where density-dependent recruitment effects have been observed (2000 cockles per m2; de Montaudouin and Bachelet, 1996). (Sanchez-Salazar et al., 1987) attributed shore-height-related differences in cockle size to crab preference for smaller cockles at low levels and upper-shore foraging by oystercatchers (with a preference for cockles >20 mm). These differential predation processes may be occurring in Strangford. The patchiness in age structure also suggests heterogeneity in recruitment and mortality, such that age classes can be strong in one area but absent from another.
Cockle reburial is not influenced by hand-raking. A comparison with the results of Coffen-Smout and Rees (1999) suggests that the experimental design was sufficiently powerful to have detected the magnitude of reburial effects associated with mechanical harvesting (average of 45% difference between controls and cockles subject to simulated mechanical harvesting: 10 drops onto a steel plate). Although raking exposes cockles on the surface, observations by Coffen-Smout and Rees (1999) indicate that suspension and reburial are part of the normal dynamics of cockle beds (with up to three cockles arriving in cleared plots every 14 d).
The impacts on Zostera are more significant for the conservation status of Strangford Lough. The experimental treatments reduced Zostera biomass by 88%. Cockle harvesting is often restricted to the summer to improve the harvested biomass and to reduce conflict with over-wintering birds. Summer cockle-raking in Zostera beds could, therefore, make a large difference to the food available to over-wintering geese when they arrive in autumn (Portig et al., 1994). The effect was similar in two seagrass bed types: Z. noltii dominated (Ringneil), and a mixed bed of Z. noltii and Z. angustifolia (North End). Cockle digging rather than raking has similar impacts on Zostera marina biomass in Yaquina Bay, Oregon (Boese, 2002). Our conclusions differ from those of Boese (2002) because of the longer period over which we demonstrated an impact on Zostera biomass, the presence of Z. marina, and the greater biomass in the Yaquina Bay plots. These factors may have led to more rapid recovery being found by Boese (2002) than in the intertidal Z. noltii/Z. angustifolia beds in Strangford.
Although the rate of recovery from hand-raking can be highly variable in space and time, a low intensity of traditional harvesting appears to have little impact on benthic communities (Kaiser et al., 2001), over-wintering birds (Stillman et al., 2001), or rates of cockle reburial. A management plan for Strangford could be as simple as to restrict cockle gathering from areas where Zostera beds prevail. This level of plan has the benefit of simplicity in that it is easy to communicate and justify to conservationists and harvesters. There are likely to be benefits too in recognizing the sustainability of traditional cockle harvesting in an SAC. These could include improving the acceptance of the designation among local communities and improving the status of harvesters such that they could become more engaged with the management aims for the Lough. Such an approach could be augmented by a light touch monitoring programme, in which stocks in accessible areas are sampled and the price of cockles is tracked. Assuming that increases in harvesting will be driven by the economic yield of hand-raking, more onerous management (e.g. licensing) could be triggered if high densities of cockles and high prices coincide. With the number of planned SACs and MPAs worldwide increasing over time, intensive management is unlikely to be a practical option for regulatory agencies. Development of a model for economic yield may therefore be a useful tool in assessing spatial and temporal changes in the potential impacts of traditional shellfish harvesting.
|
Acknowledgements |
|---|
EM was supported by a Department of Agriculture and Rural Development, Northern Ireland, Postgraduate Studentship Award. Survey work was part-funded by Quercus, a partnership between the Environment and Heritage Service and Queen's University Belfast.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Anon. (2000) Culture, Arts and Leisure Committee Report on The Fisheries (Amendment) Bill (NIA Bill 9/99). ISBN 0 33 960018 7. Available online at: http://www.niassembly.gov.uk/culture/culturereport.htm.
Boese B. L. (2002) Effects of recreational clam harvesting on eelgrass (Zostera marina) and associated infaunal invertebrates: in situ manipulative experiments. Aquatic Botany 73:63–74.[CrossRef]
Castilla J. C. and Defeo O. (2001) Latin American benthic shellfisheries: emphasis on co-management and experimental practices. Reviews in Fish Biology and Fisheries 11:1–30.
Coffen-Smout S. S. and Rees E. I. S. (1999) Burrowing behaviour and dispersion of cockles Cerastoderma edule following simulated fishing disturbance. Fisheries Research 40:65–72.[CrossRef]
de Montaudouin X. (1996) Factors involved in growth plasticity of cockles Cerastoderma edule (L.) identified by field survey and transplant experiments. Journal of Sea Research 36:251–265.
de Montaudouin X. and Bachelet G. (1996) Experimental evidence of complex interactions between biotic and abiotic factors in the dynamics of an intertidal population of the bivalve Cerastoderma edule. Oceanologica Acta 19:449–463.
Environment Agency. (2001) Regulating Order under section 1 of the Sea Fisheries (Shellfish) Act 1967. for cockles (Cerastoderma edule). Available online at: www.environment-agency.gov.uk/commondata/acrobat/deeman_plan_1263651.pdf.
European Council Directive 92/43/EEC. (1992) On the conservation of natural habitats and of wild fauna and flora. Official Journal of the European Union L. 206 (22.07.1992).
European Council. (2004) Adopting, pursuant to Council Directive 92/43/EEC, the list of sites of Community importance for the Atlantic biogeographical region. Official Journal of the European Union L. 387/1 (29.12.2004).
Gubbay S. and Knapman P. A. (1999) A review of the effects of fishing within UK European marine sites. English Nature (UK Marine SACs Project) 134.
Jensen K. T. (1992) Dynamics and growth of the cockle, Cerastoderma edule on an intertidal mud-flat in the Danish Wadden Sea: effects of submersion time and density. Netherlands Journal of Sea Research 28:335–345.[CrossRef]
Kaiser M. J., Broad G., Hall S. J. (2001) Disturbance of intertidal soft-sediment benthic communities by cockle hand raking. Journal of Sea Research 45:119–130.[CrossRef]
Legendre P. and Legendre L. (1998) Numerical Ecology. (Elsevier, Amsterdam)853.
Lotze H. K. and Milewski I. (2004) Two centuries of multiple human impacts and successive changes in a North Atlantic food web. Ecological Applications 14:1428–1447.
MacKenzie C. L. and Pikanowski R. (2004) Gear effects on marine habitats: harvesting northern quahogs in a shallow sandy bed at two levels of intensity with a short rake. North American Journal of Fisheries Management 24:1221–1227.[CrossRef]
Mistri M., Cason E., Munari C., Rossi R. (2004) Disturbance of a soft-sediment meiobenthic community by clam hand raking. Italian Journal of Zoology 71:131–133.
Nautilus. (2000) Study into Inland and Sea Fisheries in Wales. Report to National Assembly of Wales. 125 Available online at: www.nautilus-consultants.co.uk/pdfs/wales.pdf.
Piersma T., Koolhaas A., Dekinga A., Beukema J. J., Dekker R., Essink K. (2001) Long-term indirect effects of mechanical cockle-dredging on intertidal bivalve stocks in the Wadden Sea. Journal of Applied Ecology 38:976–990.[CrossRef]
Portig A. A., Mathers R. G., Montgomery W. I., Govier R. N. (1994) The distribution and utilization of Zostera species in Strangford Lough, Northern-Ireland. Aquatic Botany 47:317–328.[CrossRef]
Richardson C. A., Crisp D. J., Runham N. W., Gruffydd L. D. (1980) The use of tidal growth bands in the shell of Cerastoderma edule to measure seasonal growth-rates under cool temperate and sub-arctic conditions. Journal of the Marine Biological Association of the UK 60:977–989.
Roberts D., Davies C., Mitchell A., Moore H., Picton B., Portig A., Preston J., et al. (2004) Strangford Lough Ecological Change Investigation (SLECI). (Report to Environment and Heritage Service by the Queen's University, Belfast) Available online at: http://www.ehsni.gov.uk/pubs/pubs.shtml.
Salz R. J. and Loomis D. K. (2004) Saltwater anglers' attitudes towards marine protected areas. Fisheries 29:10–17.
Sanchez-Salazar M. E., Griffiths C. L., Seed R. (1987) The interactive roles of predation and tidal elevation in structuring populations of the edible cockle Cerastoderma edule. Estuarine Coastal and Shelf Science 25:245–260.[CrossRef]
Stillman R. A., Goss-Custard J. D., West A. D., Durell S. E. A. L. D., Mcgrorty S., Caldow R. W. G., Norris K. J., et al. (2001) Predicting shorebird mortality and population size under different regimes of shellfishery management. Journal of Applied Ecology 38:857–868.[CrossRef]
West A. D., Goss-Custard J. D., McGrorty S., Stillman R. A., Durell S. E. A. L. V., Stewart B., Walker P., et al. (2003) The Burry shellfishery and oystercatchers: using a behaviour-based model to advise on shellfishery management policy. Marine Ecology Progress Series 248:279–292.
Wood L. J. (2005) MPA Global: a database of the world's marine protected areas. Modified version for inclusion on WDPA CD. Sea Around Us Project UNEP-WCMC and WWF www.mpaglobal.org.
Zar J. H. (1996) Biostatistical Analysis. 3rd edn. (Prentice-Hall, London)662.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||