1, Coastal salt-marshes in Great Britain are of both national and international significance for their breeding redshank Tringa totanus populations. These habitats are currently under threat, as a result of habitat loss due to erosion and changes in the use of salt-marshes for grazing livestock. 2, The application of 'soft engineering' methods to coast protection and sea defence work, and the development of management plans for estuaries, offer mechanisms to create and manage salt-marsh habitats suitable For breeding redshank. However, there is a need for more information on the importance of different salt-marsh habitats and their grazing management, which could be used to ensure that these various mechanisms benefit the conservation of redshank in Britain. 3. The primary aim of our study was to describe, in detail, the relationship between the density of breeding redshank, the extent of different salt-marsh habitats and their grazing management. Our approach consisted of undertaking a survey of redshank breeding on a sample of 50 salt-marsh plots from 19 sites located in the Wash in eastern England. We used standard survey methods during which we also collected detailed data on salt-marsh habitats, vegetation structure and the intensity of grazing from survey sites. 4, Multiple regression modelling revealed that redshank densities were positively correlated with the extent of the sea-couch grass community on survey plots, but the form of this relationship varied in relation to grazing intensity. On heavily grazed plots, breeding: density increased significantly more rapidly as the extent of the sea-couch grass community increased, compared with ungrazed plots. The rate of increase on moderately grazed plots was intermediate. Breeding densities were also negatively correlated with the extent of the glasswort and annual sea-blite community, although the form of this relationship appeared unaffected by the intensity of grazing. Redshank densities also varied significantly between sites, in addition to any variation attributable to habitat or grazing. 5. Our analyses suggested that the relationship between the extent of the sea-couch grass community and redshank density, under different grazing intensities, could be explained by variation in vegetation structure. Heavily grazed plots, dominated by the sea-couch grass community, supported the most structurally diverse vegetation, and the highest breeding densities. In contrast, ungrazed plots of similar habitat contained tall, uniform vegetation and supported significantly lower breeding densities. 6, Our analyses indicate that grazing salt-marsh sites dominated by sea-couch grass on the Wash is beneficial to breeding redshank. This means that current grazing practices should be maintained at such sites and grazing re-introduced to sites which were formerly grazed, The density of cattle on the Wash is approximate to 1 cow per hectare of grazed salt-marsh, suggesting that this level of grazing pressure would be most appropriate. Caution is required, however, if sites dominated by lower salt-marsh habitats are grazed, because any slight increase in breeding density resulting from grazing might be offset by an increased risk of nest loss due to trampling. Tn this case, cattle should, if possible, be put on to the marsh towards the end of the redshank nesting season (i,e. late May/early June) to minimize this trampling risk.
