Background

Lowland heath is a priority habitat for nature conservation because it is a rare and threatened habitat supporting a characteristic flora and fauna. Only about 70,000 ha of lowland heath now remains in the UK, representing approximately 16% of its extent in the 19th century. Many heaths have been lost due to afforestation, agricultural conversion and development (JNCC 2004). The main threat today is percieved to be encroachment of trees and scrub and changes in vegetation structure resulting from lack of appropriate management such as selective grazing, controlled burning and cutting (UK BAP 2005). Nutrient enrichment, particularly deposition of nitrogen compounds emitted from intensive livestock farming, or from other sources, fragmentation and disturbance from developments such as housing and road constructions and agricultural improvement including reclamation and overgrazing, (especially in Northern Ireland) are also considered as current threats (UK BAP 2005).

The disappearance of grazing from lowland heath in the 20th century is often cited as the major cause of loss of heath vegetation in Europe (Harrison 1976, Bunce 1989, Webb 1990, Marrs 1993, Bullock & Pakeman 1996). Cessation of grazing allows succession towards woodland to proceed in the absence of alternative management (Bullock & Pakeman 1996, Manning, Putwain & Webb 2004). Conversely, grazing with livestock prevents succession, controlling scrub and maintaining open short dwarf shrub stands with a high diversity of low growing species thus achieving conservation management objectives (Gimingham 1972, 1992, Webb 1986, Bullock & Pakeman 1996, Lake, Bullock & Hartley 2001). However, the need for introducing or re-instating grazing of lowland heaths has been challenged in England and Wales, particularly (but not exclusively) where fencing of heathland on common land is required for control of grazing stock and also when other forms of management are undertaken (UK BAP 2005). There is therefore a requirement to determine how the impact of grazing compares with other management methods in delivery of conservation objectives.

Management Options

There are three primary alternatives to grazing namely no management, burning and cutting. Other forms of management e.g. turf stripping are practiced but are not generally used extensively. As already stated lack of management results in succession towards woodland and is generally thought to result in a reduction in conservation value (Bullock & Pakeman 1996, Lake, Bullock & Hartley 2001, Manning, Putwain & Webb 2004).

Burning has agricultural benefit and is usually used to increase the quality of forage available to stock (Lake, Bullock & Hartley 2001). Frequent burning especially in combination with grazing may have a deleterious impact on conservation value particularly to bryophyte and lichen-rich heathland (JNCC 2004). Controlled burning removes most above-ground living biomass, but leaves the litter layer largely intact. Thus it creates areas bare of vegetation which are re-colonised mostly by species that can resprout from underground organs. It also can cause net loss of nutrients in smoke and in run-off (Lake, Bullock & Hartley 2001). The aim of burning is to remove degenerate heather growth and to create a mosaic of different aged heather stands (Lake, Bullock & Hartley 2001).

Cutting simply removes all vegetation to a uniform height. It is a less severe management than burning, but again removes degenerate shrub growth, opens up the vegetation, and can be used to create a patchwork of stands of different heights (Lake, Bullock & Hartley 2001). The impacts of grazing management have rarely been compared within the same heath system to those of other management practices, but Pywell et al. (1995) found large vegetation differences among different areas managed in different ways (Lake, Bullock & Hartley 2001). In particular grazed areas had greater botanical species richness than areas mown or recovering from burns. The former had a higher incidence of low-growing and small forbs and grasses (Lake, Bullock & Hartley 2001). This is consistent with the current consensus on lowland heath management which suggests that burning and mowing cannot provide the diversity created by grazing (Gimingham 1972, 1992, Webb 1986, Lake, Bullock & Hartley 2001). Conversely, grazing alone is unlikely to maintain a lowland heath habitat in favourable condition. The precise impacts of grazing are variable partly because of the large number of effect modifiers that operate in grazing systems.

Reasons for heterogeneity: Potential effect modifiers

Grazing intensity, grazing period, grazing duration, stock type (including breed, gender, age, origin and husbandry), initial floristic composition, follow up period and the amount and proximity of palatable grassland within the heathland mosaic have been identified as important potential effect modifiers in upland (Gimingham 1972, Armstrong & Milne 1995, Milne et al. 1998) and lowland (Gimingham 1972, 1992, Webb 1986, Lake, Bullock & Hartley 2001) heath grazing systems.

Grazing intensity has a large impact on biomass offtake and selectivity which are both important determinants of floristic composition (Armstrong & Milne 1995, Milne et al. 1998, Lake, Bullock & Hartley 2001). The nutritional characteristics of swards vary seasonally thus grazing period also effects selectivity, whilst the duration of grazing relates to the magnitude of changes due to grazing intensity and period (Lake, Bullock & Hartley 2001). Selectivity, biomass offtake, and flock behaviour vary with stock type and breed influencing the outcome of grazing interventions (Armstrong & Milne 1995, Milne et al. 1998, Lake, Bullock & Hartley 2001). The initial floristic composition represents the baseline from which changes occur, and can have a strong influence on any subsequent successional modification. Likewise, floristic responses to grazing are dependent upon time. Short-term studies can miss potentially important vegetation responses. The amount and proximity of palatable grassland within the heathland mosaic will also affect selectivity (Armstrong & Milne 1995, Milne et al. 1998). These grazing related effect modifiers operate on a range of scales creating potential variation within as well as between different grazing regimes.

Burning impacts also vary depending on the timing, extent, frequency and severity of the burn. Likewise, the timing, extent and frequency of cutting will modify the impact of cutting. Whether cut material is left on site or removed can also influence the outcome of management. The relative impact of different management regimes on lowland heath cannot be ascertained without concurrent investigation of the impact of these factors.

Current evidence and the need for systematic review

Despite the number of high quality reviews on lowland heath management (Gimingham 1972, 1992, Webb 1986, Lake, Bullock & Hartley 2001), there is little scientific literature to provide an evidence-base to support decision-making. Much of the data on the impacts of grazing on lowland heath are not easily interpretable, having been collected from sites with no baseline monitoring, from studies which were insufficiently replicated and/or with insufficient monitoring (Lake, Bullock & Hartley 2001). Likewise, there is little robust information on burning or cutting and there are very few direct head to head comparisons of different management types. A large quantity of anecdotal and observational information from site managers and unpublished reports exist, and reviews notably Lake, Bullock & Hartley (2001) provide a qualitative synthesis of this data. A rigorous quantitative synthesis designed to minimize the potential for bias would build on this work, improving the evidence base by providing best available evidence regarding the relative impacts of grazing burning and cutting on the conservation value of lowland heath.

Systematic review methodology will be used to retrieve data pertaining to the impact of grazing, burning and cutting on the floristic composition of lowland heath and associated fauna of conservation concern. The review will limit bias through the use of comprehensive searching, specific inclusion criteria and formal assessment of the quality and reliability of the studies retrieved. Subsequent data synthesis will summarise evidence guiding the formulation of appropriate evidence-based management guidelines and highlighting gaps in research evidence. The review should be of use to staff carrying out or advising on lowland heathland site management for conservation organisations (statutory and non-statutory) and local authorities, as well as agri-environment scheme advisors. It will have value across local and national scales informing local management or policy decisions but also influencing regional policy on lowland heathland management, and national options within management frameworks such as agri-environment schemes.