Changes in breeding wader populations on lowland wet grasslands in England and Wales: causes and potential solutions

Authors

  • ANDY M. WILSON,

    Corresponding author
    1. British Trust for Ornithology, The Nunnery, Thetford, Norfolk, IP24 2PU, UK
    Search for more papers by this author
    • Present address: School of Forest Resources, University Park, Pennsylvania State University, State College, PA 16802, USA.

  • MALCOLM AUSDEN,

    1. The Royal Society for the Protection of Birds, The Lodge, Sandy, Bedfordshire SG19 2DL, UK
    Search for more papers by this author
  • TIM P. MILSOM

    1. Central Science Laboratory, Department for Environment, Food and Rural Affairs, Sand Hutton, York YO41 1LZ, UK
    Search for more papers by this author

*Corresponding author. Email: amw328@psu.edu

Abstract

Populations of waders breeding on lowland wet grassland in England and Wales have declined markedly in recent decades; the loss of once widespread species such as Lapwing Vanellus vanellus, Snipe Gallinago gallinago and Redshank Tringa totanus from many areas is of particular conservation concern. These declines are due to loss of grassland to other land uses, and to significant changes in grassland management. Drainage, reseeding and changes in grazing regimes have all been particularly detrimental to lowland wet grassland in terms of a breeding habitat for waders. Careful management of key sites, many of them managed as nature reserves, has shown that wader declines can be halted or even reversed. Aspects of this management can be applied to larger areas through agri-environment schemes, such as the Environmentally Sensitive Areas (ESA) scheme, Tir Gofal (in Wales) and Countryside Stewardship (in England) but these need be carefully targeted to ensure that the benefits for waders are maximized. In particular, it has been shown that higher tier management options within the ESA scheme (those that enhance the landscape) are more cost-effective than lower tier options (those that maintain the landscape). The extent and suitability of lowland wet grassland will face further pressure in years to come as a result of climate change, the impacts of which need to be assessed and mitigated against.

Lowland wet grassland comprises pastures and hay meadows, which are periodically flooded or which overlie waterlogged soils, and which lie below c. 200 m altitude (Jefferson & Grice 1998). The habitat supports a distinctive assemblage of birds and is well known as an important breeding habitat for waders. For seven species of wader, lowland wet grasslands hold significant breeding populations, at least in part of their UK range: Oystercatcher Haematopus ostralegus, Lapwing Vanellus vanellus, Ruff Philomachus pugnax, Snipe Gallinago gallinago, Black-tailed Godwit Limosa limosa, Curlew Numenius arquata and Redshank Tringa totanus. Of these, five are relatively numerous and are found in a range of habitats in the UK and are considered wider countryside species (O’Brien & Bainbridge 2002), whereas Ruff and Black-tailed Godwit are rare and virtually confined to lowland wet grassland.

For many years there has been concern for the conservation status of most of these wader species, both in the UK and across their European range, which in recent years has become more urgent. Black-tailed Godwit is Red-listed in Birds of Conservation Concern due to a historical decline in its UK breeding population (Gregory et al. 2002). Five species are Amber-listed: Ruff, due to the small UK population size; Lapwing, Snipe and Redshank due to their breeding population declines of 25–50% in 25 years, and Curlew because the UK supports more than 20% of the European breeding population (Gregory et al. 2002). With the exception of Oystercatcher, all of these grassland waders are in decline in substantial parts of their European range (Hagemeijer & Blair 1997) while Black-tailed Godwit and Redshank are Species of European Conservation Concern owing to their declining breeding populations (Heath et al. 2000).

WADER POPULATION CHANGES

Historically, breeding waders on lowland wet grassland in the UK have experienced mixed fortunes which have been intrinsically linked to changes in agricultural practice. Declines were noted for five wader species during 19th century (Holloway 1996) but agricultural depression in the early decades of the 20th century led to more favourable conditions, benefiting Lapwing, Snipe and Redshank (O’Connor & Shrubb 1986). During this time, both Snipe and Redshank recolonized much of lowland England while Curlew spread from the uplands into lowland counties for the first time (Alexander & Lack 1944). However, during the second half of the 20th century, there was a phase of sustained agricultural intensification including increased drainage and marked changes in grassland management practices, which together with increasing river regulation and impoundment have been associated with widespread decreases of grassland wader populations (Smith 1983, Gibbons et al. 1993, Wilson et al. 2005).

Declines were thought to be well advanced by the early 1980s, prompting the first national survey. In 1982, an extensive survey of lowland wet grassland in England and Wales (Smith 1983) revealed that populations of waders were low in many areas, and that a large proportion of waders was found on a small proportion of lowland wet grassland. A repeat sample survey in 1989 (O’Brien & Smith 1992) suggested large declines but was hampered by the limited sample of sites covered. A repeat survey in 2002 (Wilson et al. 2005) revealed further declines in four wader species in most parts of England and Wales (Table 1), with significant overall decreases of 37% for Lapwing, 61% for Snipe, 38% for Curlew and 29% for Redshank. During the same period Oystercatcher increased significantly by 51%.

Table 1.  Size of breeding wader populations on lowland wet grasslands in England and Wales in 2002, and population changes since 1982.
 EnglandWales
Pairs 2002Change (%) 1982–2002(95% CI)Pairs 2002Change (%) 1982–2002(95% CI)
  1. Source: Wilson et al. (in press).

Oystercatcher 959+52(+27 to +81)  8+17 (−80 to +367)
Lapwing5266−37(−47 to −26)121−69  (−87 to −46)
Snipe 558−62(−75 to −50) 12 +9(−100 to +217)
Curlew 412−34(−49 to −26) 29−76  (−91 to −38)
Redshank2433−20(−10 to −44) 19−62(−100 to +61)
No. of sites surveyed    986  65
(area in hectares) (13 840) (1150)

Ruff and Black-tailed Godwit are now scarce breeding birds in the UK, found mainly on lowland grassland nature reserves in southeast England. Data supplied to the Rare Breeding Birds Panel (Spencer et al. 1985, 1991, Ogilvie et al. 2003) suggest that the number of confirmed Ruff nests or broods ranged from 0 to 7 per year between 1973 and 2001 while the number of breeding localities declined during the 1990s (Fig. 1). Black-tailed Godwit numbers declined during the 1970s and 1980s but there has been a steady recovery since then (Fig. 1).

Figure 1.

Trends in Ruff and Black-tailed Godwit numbers in the UK between 1973 and 2001. Sources: Spencer et al. (1985), Spencer et al. (1991) and Ogilvie et al. (2003).

CONSERVATION STATUS OF LOWLAND WET GRASSLAND

In part for their importance for breeding and non-breeding waterfowl, lowland wet grasslands are encompassed within the UK Biodiversity Action Plan (BAP) under three priority habitat types:

All of these habitats have declined substantially in extent; for example, there is currently estimated to be 300 000 ha of grazing marsh nationwide, following a loss of more than 40% since 1930, primarily though drainage and agricultural improvements (http://www.ukbap.org.uk). Much of the remaining lowland wet grassland currently holds no breeding waders. For example, in a survey of 1051 key sites in 2002, 90% of Snipe were found on just 3% of the grassland surveyed (Wilson et al. 2005), despite the fact that survey sites were initially selected on their suitability for waders (Smith 1983). For some species, the reduction in site occupancy has been marked; Snipe, for example, was found on 25% of sites surveyed in 1982 but just 8% in 2002 (Wilson et al. 2005). This loss of birds from large tracts of lowland wet grassland confers great significance on those sites that do still hold significant concentrations of breeding waders. Many of the most important sites are already protected and managed as nature reserves.

HABITAT REQUIREMENTS OF GRASSLAND WADERS AND REASONS FOR DECLINES

Breeding Lapwing, Redshank, Snipe and Curlew on lowland wet grassland are often associated with less intensively managed swards and high water levels (Green & Robins 1993, Vickery et al. 1997, Milsom et al. 2000, 2002a, Henderson et al. 2002, Wilson et al. 2005). They also prefer open, undisturbed areas, avoiding hedgerows, pylons and roads (Reijnen et al. 1996, Milsom et al. 2000). Their specific requirements differ in their preferred ranges of sward height for nesting and foraging, and in how water levels affect the abundance and accessibility of their invertebrate prey.

Adult Lapwings and Redshank feed on soil invertebrates among grass swards and shallow pools; Redshank on coastal grazing marsh will also fly to nearby estuaries to feed (Beintema et al. 1991, Johansson & Blomqvist 1996, Ausden et al. 2003). Both Lapwings and Redshank thus benefit from retention of shallow floods during the breeding season, particularly once they have chicks (Ausden et al. 2003). Lapwings tend to prefer shorter swards than Redshank (Green 1986, Milsom et al. 2000), whereas Snipe prefer taller, more tussocky grassland than both Lapwings and Redshank (Mason & MacDonald 1976). Snipe probe the soil to feed on macroinvertebrates, thus requiring soft soil (Green 1988, Green et al. 1990), which can be maintained by extensive flooding. However, this has the disadvantage of reducing prey biomass (Ausden et al. 2001).

Breeding Black-tailed Godwits within the UK are largely confined to two peat washlands: the Ouse and Nene Washes in eastern England, although small numbers also breed on coastal grazing marsh. Presence of breeding Oystercatchers on lowland wet grassland has been found to be positively associated with cattle grazing and variable field topography and negatively associated with the presence of tussocky vegetation (Milsom et al. 2000, Wilson et al. 2005). On upland grassland Oystercatchers appear to benefit from agricultural improvement, in contrast to other breeding wader species (Baines 1988). There is little specific information on the habitat requirements of the Curlews or Ruff on lowland wet grassland.

Conversion of lowland wet grassland to arable (e.g. Williams et al. 1983, Williams & Hall 1987, Dargie 1993) has reduced the area of grassland available to breeding waders during the 20th century. However, during the period of breeding wader declines since the early 1980s, there has been little further overall loss of grassland to arable (Wilson et al. 2005). Instead, wader declines are thought to be due primarily to the intensification of agricultural management, such as re-seeding, use of artificial fertilizers and switch from hay to silage, and lowering of water levels, which decreases the suitability of grassland for most breeding wader species (e.g. Green & Robins 1993).

Widespread agricultural intensification has occurred since the 1970s (e.g. Robinson & Armstrong 1988, Vickery et al. 2001), which has an impact upon the habitat requirements of waders, specifically through changes in sward structure and the increase in area of silage. Fields cut for silage tend to have a dense, tall and uniform sward in spring compared with hay meadows or cattle-grazed pasture (Vickery et al. 2001, Atkinson et al. 2004), making them less suitable for Lapwings (e.g. Milsom et al. 1998, 2000). Silage is also cut earlier in the year than hay, and this increases the proportion of nests and chicks that are lost during mowing (Beintema & Müskens 1987, Kruk et al. 1996) while loss of clutches during agricultural operations tends to be far higher on improved than unimproved grassland (Baines 1990). Agricultural improvement is also accompanied by a decrease in the mean size of arthropods (Blake et al. 1994), which may decrease the profitability of this prey item for wader chicks and thereby reduce chick survival (Beintema et al. 1991).

Agricultural improvement has also led to increased stocking densities, which then increases the proportion of nests lost through trampling (Beintema & Müskens 1987, Green 1988). An increase in sheep densities also reduces the heterogeneity of the sward (Vickery et al. 2001), and thereby reduces the availability of tussocky grassland preferred by nesting Redshank and Snipe (Mason & MacDonald 1976, Green 1986, Milsom et al. 2000).

The effects of predation on population changes in breeding waders on lowland wet grassland are poorly understood. Although losses of nests and chicks attributed to predation can be high (Beintema & Müskens 1987, Green et al. 1987b, Baines 1990, Grant et al. 1999), there have been no published studies on the effects of predator removal on hatching success or post-breeding populations, although a number are underway. There is evidence that clutch losses due to predation on upland grassland increase following agricultural improvement, probably due to the reduced heterogeneity of the improved swards making nests more conspicuous to predators (Baines 1990, Whittingham & Evans 2004). In addition, populations of Lapwings that have been reduced by habitat-related factors are likely to suffer increased levels of predation, due to a reduction in their ability to defend communally their nests and eggs against predators (Elliot 1985, Berg 1996, Seymour et al. 2003).

Lowland wet grassland breeding waders will replace clutches lost early in the breeding season. Therefore, prolonging the period during which suitable breeding conditions are present can increase the likelihood of these birds re-nesting and successfully raising chicks (Beintema & Müskens 1987, Green 1988).

POTENTIAL SOLUTIONS

Creation of appropriate conditions

Land managers have a variety of habitat management techniques at their disposal to enhance the quality of existing grassland for breeding waders. These include manipulating water levels and surface wetness in spring, regulating livestock grazing during the breeding season, suspending agricultural operations during critical periods to minimize the risk of nest destruction, and targeting the foregoing measures at fields where the birds are most likely to respond positively.

Maintenance of high water tables until mid-summer is desirable to ensure that soils remain soft, and to provide pools where wader broods may feed on aquatic prey (Green 1988, Ausden et al. 2003). In principle, this could be achieved by raising water levels in adjacent ditches, using the methods described by Self et al. (1994), to facilitate subsurface irrigation of the upper soil layers. However, the practicability of this approach depends upon the hydraulic conductivity of the soils, which are sufficiently high in peat but negligible in marine clays (Armstrong 2000, Gavin 2003). On coastal grazing marshes, which overlie marine clays, the only practicable means of retaining surface wetness is to flood the grassland. Then the resultant pattern of flooding depends upon surface topography (Milsom et al. 2002a, 2002b). On flat fields, sheet flooding is likely which, if it is prolonged, often reduces the abundance of soil invertebrate prey, especially earthworms (Ausden et al. 2001). It also results in sward death, thereby reducing the value of the grassland for livestock grazing or hay cropping (Mountford et al. 1997). A possible remedy to these problems has been identified by the RSPB, who have demonstrated the practicability of excavating surface drainage channels, using a rotary ditching machine, to create localized floods on sites that lack natural channels (Collings 2002, Lambert 2003).

Although grazing is essential to produce swards of the requisite height and structure, the presence of livestock may decrease breeding productivity by nest trampling or disturbance (Beintema & Müskens 1987, Green 1988, Hart et al. 2002). These effects can be either mitigated by reducing stock densities during critical periods or eliminated by excluding livestock from selected fields that support high densities of birds. The latter option is practicable where the rate of grass growth is sufficiently slow in spring, as in the North Kent Marshes, for swards to remain suitable for Lapwings during their nesting period (Hart et al. 2002).

Large areas of lowland grassland were converted to arable during the second half of the 20th century. Re-creation of lowland wet grassland from arable land is a target in the Habitat Action Plan for Floodplain and Coastal Grazing Marsh (UK Biodiversity Steering Group 1995). Such re-creation has been adopted as a component of agri-environment schemes, resulting in some extensive reversion to grassland, notably in Kent, Sussex and the Norfolk Broads.

The role of agri-environment schemes and site protection

There are three complementary mechanisms for conserving breeding waders in England and Wales: (1) management of nature reserves, (2) Sites of Special Scientific Interest (SSSI) legislation and (3) compliance with agri-environment scheme management prescriptions. These mechanisms are not mutually exclusive. For example, an SSSI might be managed as a nature reserve and entered into an agri-environment scheme management agreement.

Nature reserves contain only a small proportion of lowland wet grassland but contain the majority of breeding Black-tailed Godwits and lowland wet grassland breeding Snipe. Breeding waders have fared better on nature reserves than on non-reserves (Ausden & Hirons 2002, Wilson et al. 2005), although reserve management has been relatively unsuccessful at increasing numbers of breeding Snipe, possibly mainly because of lack of suitable peat soils, which are generally favoured by this species. In England only 2.6% of lowland wet grassland, for which the soil type has been classified, is on peat (Dargie 1993).

Most of the important lowland wet grassland sites for breeding waders are designated SSSIs. Legislation under the Wildlife & Countryside Act 1981 (as amended) now requires landowners of SSSIs to put in place all necessary management measures required to maintain or achieve favourable conservation status of the features for which the SSSI was notified. This has the potential to benefit breeding waders at sites for which they are a notified feature. However, O’Brien and Smith (1992) showed that the SSSI mechanism did not deliver up to 1989, although some sites were not notified for breeding waders and management was targeted on other conservation interests. Recent strengthening of the legislation concerning SSSIs in England, following the Countryside & Rights of Way Act 2000, should facilitate the maintenance of wader populations on designated sites in the future.

Conservation of waders in the wider countryside can only be achieved through successful implementation of agri-environment schemes. There are three such schemes in England and Wales that offer payments to landowners to manage lowland wet grassland in ways aimed at benefiting breeding waders. These are Tir Gofal in Wales, Countryside Stewardship (CS) in England and Environmentally Sensitive Areas (ESAs) in both countries (Coates 1997, Swash 1997). A new integrated Higher Level Scheme (HLS) combining CS and ESAs will be introduced in 2005. ESAs have the potential to make a major contribution to the conservation of breeding waders, as their boundaries were determined with reference to lowland wet grassland. As a result, they cover eight of the largest and least fragmented tracts of lowland wet grassland remaining in England and Wales (Tallowin & Mountford 1997). Farmers within ESAs may enter into agreements with Defra, which require them to follow management prescriptions that are designed to maintain or enhance characteristic plant and animal communities of lowland wet grassland. In return, agreement holders receive compensatory payments. The prescriptions are tiered according to the demands made on the agreement holder. Prescriptions in lower tiers are largely aimed at maintaining the landscape, whereas prescriptions in the higher tiers involve enhancements, and as such, attract higher payments. Two groups of prescriptions are aimed specifically at breeding waders: restrictions on stocking rates in spring and raised water levels.

How successful have ESAs been at conserving breeding wader populations on lowland wet grassland? In the Norfolk Broads ESA, wader numbers increased where water levels had been raised but declined on other land under ESA agreement (O’Brien & Self 1994). A similar pattern was reported from the Somerset Levels and Moors ESA (Glaves 1998). A more recent analysis of population trends in four ESAs (Norfolk Broads, Somerset Levels and Moors, Suffolk River Valley and Upper Thames Tributaries) revealed declines in the numbers of Lapwing, Redshank and Snipe since each ESA had been designated (Ausden & Hirons 2002). In the Norfolk Broads, Somerset Levels and Moors, and the Suffolk River Valleys, numbers of Lapwing and Redshank declined or remained steady on non-reserve land but on RSPB reserves within the three ESAs, numbers had increased over the same period (Ausden & Hirons 2002). The implication was that the habitat management applied in nature reserves had succeeded in stabilizing or reversing population declines, whereas the prescriptions applied on land elsewhere within the ESAs had not been as effective. The findings of this analysis were largely confirmed by a comparison of changes in numbers over the longer term (1982–2002) in the key ESAs for breeding waders (Wilson et al. 2005). Wader numbers have declined in six of these eight ESAs, the North Kent Marshes and Essex Coast ESAs being the outstanding exceptions. It should be noted that the ESAs were designated after the 1982 survey, between 1987 and 1993 (Swash 1997).

A number of factors may have contributed to the apparent failure of most ESAs to protect breeding wader populations outside of nature reserves. In a review of the effectiveness of agri-environment schemes in The Netherlands, Kleijn et al. (2001) identified three possible causes: (1) that conservation management prescriptions developed under experimental conditions may not deliver the expected benefits in practice, (2) that the prescriptions may have unintended side-effects and (3) that socio-economic factors may inhibit the uptake of prescriptions (e.g. wet grassland may be less attractive to farmers as they complicate livestock husbandry and may reduce forage quality). Two additional possibilities are (4) inappropriate or lack of targeting of prescriptions, and (5) failure by participants in agri-environment schemes to implement prescriptions (compliance).

Given recent experience on nature reserves (Ausden & Hirons 2002), where the recommended prescriptions are applied rigorously (1) is unlikely but (2) cannot yet be ruled out. Numbers of waders may have increased in reserves because breeding productivity is higher there than in populations on non-reserve land in ESAs. However, an alternative possibility is that the habitat created on reserves attracts birds to nest but these populations are not self-supporting because of poor breeding productivity and they rely upon immigration to maintain numbers, which might explain the declines on non-reserve land. This would not, however, cause the large and long-term declines in wader populations, which took place long before most of these nature reserves were established. As the requisite information on breeding productivity is lacking, it is currently not possible to distinguish between the two scenarios. More information is needed on wader productivity and meta population dynamics.

The declining wader population trends on ESA land outside of nature reserves may also have been partially due to socio-economic factors (3) inhibiting the uptake of raised water level prescriptions (Ausden & Hirons 2002). For example, sward death arising from prolonged sheet flooding may have acted as a disincentive to livestock farmers, as might the cost of introducing appropriate water-level management systems. Raised water-level management usually requires the creation of isolated hydrological units, often necessitating the agreement of several landowners (e.g. Jenman & Kitchin 1998), which might not always be forthcoming. It is interesting to note that ESAs with larger areas under grassland ‘enhancement’ prescriptions (higher tiers) have generally fared better then those where a greater proportion of land is under the lower tier ‘maintenance’ prescriptions (Fig. 2). These relationships do, however, require a more detailed analysis to tease out the possible effects of initial wader population levels and length of time that grassland has been in the various prescriptions. There is currently no published information on targeting of prescriptions on lowland wet grasslands (4) or compliance (5). It should also be noted that high water tables are not always compatible with management for other interests, such as botany, giving rise to potential conflicts of conservation interests.

Figure 2.

Changes in wader populations on lowland wet grassland ESA between 1982 and 2002 and percentage of land in prescriptions that was in higher tiers in 2002.

There have been no published large-scale studies of the effectiveness of prescriptions available within Tir Gofal or CS in maintaining wader populations, although some are underway. These have the potential to provide similar grassland conditions to those that can be achieved under ESA prescriptions and could be important in maintaining wader populations in the wider countryside. In order that lowland wet grassland waders are retained throughout their range, and that they may have the opportunity to re-colonize areas from which they are now lost, it is vital that agri-environment scheme options continue to be available to landowners throughout the country.

CONCLUSIONS

There is now much evidence for a long-term decline in breeding wader populations on lowland wet grassland throughout England and Wales. These declines are intrinsically linked to drainage and other changes in grassland management. The three main mechanisms for halting or reversing these declines are the establishment of more lowland wet grassland nature reserves and increasing the effectiveness of agri-environment schemes and SSSI management. The former may be limited in scope owing to financial constraints but targeting of currently unprotected key sites for purchase or expanding existing areas could be a useful way of safeguarding the important wader populations that remain. It is also vital that sites designated as SSSIs are managed appropriately to ensure that breeding wader populations within these areas are maintained or returned to former levels.

In order that maximum benefit is achieved from agri-environment schemes, it is vital that payment levels and advice given to landowners are sufficient. Evidence suggests that only the more demanding and costly management prescriptions are likely to benefit breeding waders (although there are some low-cost options for arable farmland that may benefit breeding lapwing, Sheldon et al. 2004) and therefore an appropriate balance must be struck between these and the less demanding, lower cost measures available in Tir Gofal and the new Environmental Stewardship scheme in England. There is also a need to ensure that objectives and targets are set for breeding waders, and that measures of habitat condition and numbers of breeding waders are monitored sufficiently to determine whether these schemes are proving successful or whether they need revising. It is vital to understand the links between habitat management and breeding wader productivity and how birds move between sites.

In this review we have discussed how careful management of surface water levels could be extremely beneficial for breeding wader conservation. However, one potentially important impact upon water levels that we have not reviewed is that of predicted climate change. This could result in a loss of coastal grazing marsh to managed realignment of sea defences; increased frequency of spring floods and/or drought conditions could have very detrimental impacts on habitat suitability (e.g. Green et al. 1987a) while higher rates of evapotranspiration and loss of site wetness in spring, faster and earlier grass growth and reduced length of breeding season could all have an impact in the future. Mitigation of these potential impacts should be considered and the opportunities they may bring for the creation of lowland wet grassland habitat grasped. Better design of flood defences, in particular their abilities to cope with extreme weather events, could provide suitable conditions for breeding waders. Considering the losses of breeding waders we have observed in recent years and the importance of lowland wet grassland, we should consider appropriate mitigation measures, for example the creation of replacement freshwater habitats elsewhere, long before such sites are lost.

Ancillary