United Kingdom: England: Calluna Heathland Restoration


Vegetation development after conventional bracken control with herbicide in the uplands is often slow and rarely results in the development of the target community. Two projects were conducted on contrasting sites to investigate cost-effective methods for the re-establishment of vegetation dominated by Calluna vulgaris (heather), which is the usual target community seen as the desired replacement for Pteridium aquilinum (bracken) in the uplands by many land managers. At one site, where grazing intensity was low and shelter was sufficient to prevent rapid Pteridium litter loss, the necessary treatments to establish Calluna were to disturb the litter and add a source of Calluna seeds. A difference between continued Pteridium control or a relaxation of control was not observed over the time-scale of the experiment. At the second site, where grazing intensity was high and shelter minimal, vegetation dominated by Rumex acetosella had developed on a thin but dense litter layer. To enhance Calluna regeneration it was necessary to prevent sheep grazing, disturb the remaining litter layer and add a source of Calluna propagules. A nurse crop hindered Calluna establishment due to its vigorous growth. At neither site was an initial application of fertilizer an aid to Calluna establishment. It is possible to devise methods for the re-establishment of Calluna at sites where previous attempts at Pteridium control alone have not had the desired effect on vegetation development. However, the precise choice of methods used is dependent on the grazing management at the site. High levels of livestock grazing prevented Calluna establishment, but some trampling may aid litter breakdown. The ability to adjust stocking rates and access to the site by machinery are critical factors to take into account in producing an effective strategy for long-term Pteridium control and its replacement by target vegetation.

Quick Facts

Project Location:
602 Newton Ln, Tamworth B79 0NR, UK, 52.66031495629122, -1.5841340937499808

Geographic Region:

Country or Territory:
United Kingdom



Organization Type:


Project Stage:

Start Date:

End Date:

Primary Causes of Degradation

Agriculture & Livestock

Degradation Description

In many parts of the world bracken has long been recognized as an agricultural weed. In Britain, the perception of Pteridium as a weed problem has resulted from its decline in use for agricultural, domestic, and industrial purposes, including animal bedding, thatching and soap and the apparent expansion of large areas of Pteridium into previously productive agricultural land. In is though to cover about 3700 sq km of the land surface of Great Britain, about 1.6% of the land area. Pteridium-dominated areas have little agricultural or conservation interest, except in a small number of cases, such as the occurrence of high brown and heath fritillary butterflies and their associated food plants. Pteridium is also known to invade other, more valued, plant communities including semi-natural grassland and heathland. This type of invasion is a far more potentially damaging, both for agricultural profitability, animal welfare, sporting interests and for conservation.

Reference Ecosystem Description

It is difficult to assess the pre-disturbance condition of the ecosystem, as it has had considerable human influence for hundreds of years. There has been considerable grazing pressure at these sites, with a stocking rate at the Hordron Edge site of .5 sheep/ha, while the Wetherhouse Moor site was subjected to considerable grazing pressure as part of a large common. By the time of initiation of the project the uncompacted Pteridium litter that had been originally present had been lost and the surface was composed of highly compacted litter, rhizome fragments and soil. Initial sampling of the vegetation done in 1993 at the Hordron Edge site showed that Pteridium litter comprised 99.3% of the ground cover. A small number of Chamerion angusifolium, Deschampsia flexuosa and Galium saxatile were present, as were Betula spp. Pinus sylvestris and Sorbus aucuparia. Wetherhouse Moor was sampled in 1993 as well, litter at this site made up only 80% of ground cover. The three dominants at this site were Rumex acetosella, Campylopus introflexus, and Festuca ovina.

Project Goals

The aim of this project was to develop strategis for directing succession where Pteridium control alone has not yielded results. This project was directed as restoring the vegetation dominated by Calluna, as this was the dominant species of the surrounding vegetation and the intended target of the land owner/manager at the time of the spraying.


The project does not have a monitoring plan.


There have been many attempts to control Pteridium, with two accepted methods at present: cutting and the application of the herbicide, asulam). Long-term Pteridium control is not common, except where follow-up treatments have been applied and many experimental treatments show that recovery of the Pteridium is often complete within 10 years of spraying. Aerial spraying of asulam can be cost effective for Pteridium control, particularly if followed by necessary steps to keep it in a suppressed state. The typical reason for Pteridium control is to restore other vegetation, cognizant of the fact that after aerial spraying vegetation development is slow, patchy and unpredictable. In most instances, vegetation development is often slow and tied closely to litter breakdown, while in many situations the target community is some form of Calluna vulgaris-dominated vegetation. This project was the first attempt to develop cost-effective strategies for vegetaton restoration in the marginal upland areas where the majority of Pteridium control takes place.

Description of Project Activities:
The effects of four treatments on the rate and composition of vegetation regeneration were investigated at the Hordron Edge site; (1) maintenance of Pteridium suppression, (2) litter disturbance, (3) addition of Calluna seed and (4) fertilizer addition . A split-plot design was chosen with continued Pteridium suppression or unhindered Pteridium regeneration forming the main plots. It was impractical to implement this treatment in a full factorial design, as light penetration from the side of small plots would have partly negated the shading effects of the Pteridium. Main plots were 9 m x 5 m. The other three treatments were implemented factorially in eight 1 m x 1 m subplots within each main plot. Subplots were separated from each other and the edge of the main plot by 1 m. Four, fully replicated, blocks were established in August 1993. Continued Pteridium suppression was carried out by a once-yearly cut, usually in early August. Litter disturbance was achieved by the removal of the litter layer with a rake which was carried out in October 1993. Calluna seed addition was achieved by adding 500 g of cut Calluna shoots per subplot (equivalent to 5 t/ha), again in October 1993. Fertilizer addition (150 kg/ha of ENMAG, a slow-release preparation) was carried out in April 1994, with the intention that it would benefit any recently germinated seedlings. The effects of four treatments on vegetation succession were assessed at the Wetherhouse Moor site; (1) grazing management, (2) litter disturbance, (3) addition of Calluna seed and (4) fertilizer addition. An extra level was added to the seeding treatment by inclusion of a nurse crop to aid surface stabilization and Calluna establishment. A split plot design was chosen with grazing access or exclosure forming the main plots, as it was practical to fence only four areas. Main plots were 11 m x 9 m. The other three treatments were implemented factorially in 12 subplots (1 m x 1 m) within each main plot. Subplots were separated from each other by 1 m and the edge of the main plot by 2 m. Four fully replicated blocks were established in August 1993. Livestock-proof fences were erected in October 1993. Litter disturbance, addition of Calluna seed and fertilizer addition were all carried out in the same manner and at the same rates as at Hordron Edge. The nurse crop was a mixture of Agrostis castellana (20 kg/ha) and D. flexuosa (40 kg/ha) (Emorsgate Seeds, Tilney All Saints, Norfolk). The Pteridium on this site was prevented from regeneration by once yearly cutting or pulling.

Ecological Outcomes Achieved

Eliminate existing threats to the ecosystem:
The two project sites chosen were considered to be representative of many similar sites across Great Britain. The Hordron Edge site is typical of many where a deep litter layer accumulates and ground disturbance by herbivores is minimal. The Wetherhouse Moor site is typical of sites where exposure to wind and trampling by herbivores is sufficient to remove the thick litter layer, leaving a compacted mixture of soil, litter and rhizome material as the substrate. Each of the treatments employed in the experiment had an effect on the development of vegetation at both sites. However, it should be borne in mind that the target for restoration was a community dominated by Calluna vulgaris, and analysis of the benefits of treatment effects is discussed with this aim and the constraints against successful regeneration set out above in mind. Bracken regeneration, over the course of the experiment at Hordron Edge, did not appear to have affected the growth of established vegetation. However, in the long term, it is well known that bracken will eventually shade out almost all the established vegetation by the time it has fully recovered from treatment. The deep litter layer at Hordron Edge effectively prevented the establishment of almost all vegetation, particularly as much of the moss cover was growing over relatively intact litter. Seven years after the initial spraying, a vascular plant cover of only 5.5% had established in plots where litter disturbance had not taken place. Litter disturbance increased total vegetation cover and the cover of vascular plants, especially the cover of Agrostis capillaris and Calluna, presumably by exposing bare soil for germination. Until the litter layer breaks down, it would appear that vegetation establishment and growth on this site will remain slow. Litter disturbance alone at Wetherhouse Moor, where little litter was present at the start of the experiment, had much reduced effects on the dynamics of the vegetation. To establish any Calluna cover at Hordron Edge it was necessary to add seed to areas where the litter had been disturbed. Here it contributed ca. 50% of the established cover and established plants appeared to be healthy and capable of long-term survival. In order to achieve Calluna establishment at Wetherhouse Moor it was also necessary to add a source of seed. There was little evidence of Calluna establishment from the seed bank (probably minimal) or seed rain at either site, despite the close proximity of Calluna stands. Many stands of Pteridium appear to be of considerable age and hence have impoverished seed banks necessitating seed addition. Fertilizer addition had very little effect at either site except to increase the cover of D. flexuosa, which was a natural colonist at Hordron Edge and sown as the nurse crop at Wetherhouse Moor. In fact the addition of a D. flexuosa nurse crop to stabilize the soil surface at Wetherhouse Moor had a deleterious effect on the development of Calluna cover. It would appear that the sowing of such a nurse crop, while useful in extremely nutrient poor systems and where surface stability is a problem, such as on bare peat or mine wastes, appeared detrimental where these conditions do not apply. At Wetherhouse Moor, the different vegetation dynamics within and outside the exclosures demonstrated the importance of grazing to reach the target community. The developing Calluna appeared to require protection from grazing, but protection from grazing also allowed the expansion of competitors; D. flexuosa and bracken itself. The regeneration of bracken was considerably slowed where trampling occurred. Outside the exclosures the vegetation was very different with Campylopus introflexus and Rumex acetosella dominating. This pattern of vegetation change was similar to many observed in surveys of areas subject to bracken control. Bracken regeneration appeared to be reduced by high stock levels, and D. flexuosa was an early and successful colonist of bracken litter at low grazing pressures. However, D. flexuosa also flourished on some sites where grazing sufficiently disturbed the soil surface allowing the spread of this species. Campylopus introflexus and Rumex acetosella both appeared to benefit from either reduced competition outside the exclosures, or from the continued substantial levels of disturbance. The treatments used to overcome the constraints for restoration were imposed in a factorial design, and hence interactions between treatment will guide the choice of treatment combinations. At Hordron Edge (lightly grazed, deep litter site) disturbing the litter to provide germination niches and adding seed to exploit these niches was the most successful treatment combination for Calluna establishment. This would appear to be the most likely and cost-effective method to restore a Calluna-dominated community at similar sites as long as the bracken control was continued to prevent regeneration. At Wetherhouse Moor (more heavily grazed, open ground site) Calluna cover was maximized by excluding grazing, disturbing the remaining litter and adding Calluna seed. Bracken control must also be maintained to prevent further shading of plants where protection from grazing increases the recovery of bracken from control. The broad conclusions from this project are similar to those reached in a study of vegetation restoration after bracken control on lowland heathland as well as many other studies of heathland restoration. What is clear at both sites was that in order to achieve substantial Calluna establishment and growth, seed must be added. Where any litter layer is present, this must be removed in some way (rotovation, burning, raking off) to provide suitable soil conditions for germination and establishment. However, they show that there is little benefit in allowing natural colonization to take place (unlike on lowland heathland). Intervention is necessary to produce vegetation close to the desired target. In addition, the experiments described above demonstrate a number of other important points. A nurse crop appears to hinder successful establishment and early fertilizer addition does not promote Calluna growth. Also, the success of Calluna regeneration appears to be dependent on grazing pressure, as it is prevented at high grazing intensities (Wetherhouse Moor). It appeared to be unhindered by the light grazing at Hordron Edge. However, this simple comparison demonstrates that there is a need to carry out experiments at different stocking rates. Such investigations would identify an optimal range of grazing intensity where the impact on the Calluna regeneration would be minimal, whilst maximizing the grazing of competitors and the trampling of regenerating fronds and litter. Such trade-off may not exist, depending on the grazing preference of the livestock used, and more direct management may be necessary.

Factors limiting recovery of the ecosystem:
The results of this study indicate that a strategy for restoring Calluna-dominated vegetation appears to be possible. However, feasibility and affordability must also be considered. In order to carry out many of the treatments described, access by vehicle is essential. This is clearly not an option on many sites. Litter removal can be achieved by burning, but this involves the availability of enough labour to control the fire. Seed can be spread by hand, but again this may be difficult where large areas are involved. Stock control is often difficult to achieve, because of open access on common land, the high cost of fencing or the lost profit resulting from reducing stock numbers. Thus the successful implementation of a combined bracken control and vegetation restoration program has to have in place the necessary funding to ensure an effective outcome, whether that funding comes from the landowner or from an agri-environmental scheme.

Socio-Economic & Community Outcomes Achieved

Sources and Amounts of Funding

Funding for this project came from the UK Ministry of Agriculture, Fisheries and Food.

Other Resources

Pakeman, Robin J. et al. 2000. Vegetation re-establishment on land previously subject to control of Pteridium aquilinum by herbicide. Applied Vegetation Science, 3: 95-104.

Primary Contact

Organizational Contact