Providing evidence to improve practice

Individual study: A comparison of techniques for restoring heathland on abandoned farmland at South Middlebere Heath, Dorset, England

Published source details

Pywell R.F., Webb N.R. & Putwain P.D. (1995) A comparison of techniques for restoring heathland on abandoned farmland. Journal of Applied Ecology, 32, 400-411


In the UK changes in agricultural policies have reduced the extent of cultivated farmland. This has provided opportunities to restore heather Calluna vulgaris-dominated heathland vegetation to lowland sites where it once occurred. Large-scale replicated experiments were established on abandoned farmland in southern Britain to compare the effectiveness of four treatments for heathland restoration.

Study site: Between December 1988 and April 1990 large-scale replicated experiments were carried out on abandoned farmland at South Middlebere Heath (National Grid ref: SY 964854), Corfe Castle in Dorset, southern England.

Treatments: Four treatments for heathland restoration (three replicates in 10 x 50 m plots) were compared:

i) application of herbicide - in spring 1989 vegetation in three plots was eradicated by application of the non-selective herbicide glyphosate (Roundup™) by backpack sprayer at a rate of 5 L/ha active ingredient in 200 L/ha of water. There was no rain in the 24 hours subsequent to spraying;

ii) addition of harvested heather shoots and capsules – in early December 1988 shoots were collected using a double-chop forage harvester from a nearby heathland. The upper 35 cm of the heather was removed. A rotary cultivator was used to prepare the ground prior to shoot addition. Shoots were applied at a rate of 3.3 kg/m² to a maximum depth of 4 cm and then rolled with a ring roller;

iii) addition of heathland topsoil – soil was collected from a nearby heathland stripping to a depth of 4-5 cm using a slew excavator. Prior to addition the existing grass turf was stripped to 5 cm depth. From 11-13 tonnes (22-23 kg/ m²) of topsoil was applied to each 500 m² plot. The farmland and heath soils were mixed using a rotary cultivator and rolled with a ring roller in order to achieve good contact with the seed bed;

iv) translocation of heathland turves – in July 1989, 1,500 m² of turves were collected from a nearby heathland (as topsoil site). They were dominated by characteristic heath plants e.g. C.vulgaris, cross-leaved heath Erica tetralix, bell-heather E.cinerea, gorse Ulex minor, purple moor-grass Molinia caerulea and bristle bent Agrostis curtisii. The soils was stripped to 15 cm and turves transplanted. A summer drought required watering of the turves between July and September.

v) untreated control

Monitoring and analysis: The number of seedlings of heathland plant species on each treatment was counted in December 1990 and 1991, and the shoot frequency of these species was recorded in January 1993. Samples of soil were collected and analysed in the laboratory.

Restoration success: Restoration success was assessed by comparison of the average number of seedlings (mean number/m²) of seven key heathland plant species: C.vulgaris, E.tetralix, E.cinerea, U.minor, M.caerulea, A.curtisii and heath milkwort Polygala serpyllifolia.

Soil comparison: The grassland soil had a significantly higher pH (4.9-5.2) and contained greater concentrations of extractable phosphorus (total P 0.23) and exchangeable calcium (75 >< 0-40 cm depth) than that of the adjacent heathland (pH 3.8-3.9; total P: 0.13; Ca 55.5 >< 0-40 cm depth).

Control: The controls showed that there was some natural regeneration of heathers (C.vulgaris and Erica spp.) within the grassland, although this was very poor in both years (1990: 0.8/m²; 1991: 0.2/m²). No other species were recorded.

Herbicide treatment: Spraying with glyphosate more-or-less completely inhibited all regeneration of heathland plants, with unidentified heather sp. seedlings recorded only in 1990, averaging 0.1/m². No other species were recorded.

Application of harvested heather shoots: Application of harvested heather shoots and seed capsules increased the number of seedlings of three species: C.vulgaris (1990: 44.1; 1991: 30.3); E.cinerea (1990: 9.9; 1991: 14.3); A.curtisii (0.1 in both years). However, the other four key species (i.e. E.tetralix, U.minor, M.caerulea and P.serpyllifolia) were missing.

Addition of heathland topsoil: All seven plant species occurred in greater numbers on the plots where heathland topsoil had been applied with all acceptP.serpyllifolia recorded in both years : C.vulgaris (1990: 421; 1991: 286), E.tetralix (1990: 0.1 ; 1991: 0.3), E.cinerea (1990: 11.4; 1991: 57.4), U.minor(1990: 17.9; 1991: 15.7),, M.caerulea(1990: 12.0; 1991: 10.5), A.curtisii (1990: 0.4; 1991: 1.1) and P.serpyllifolia (1991 only: 1.4).

Translocation of heathland turves: The large-scale translocation of heathland turf appeared to be feasible and instantly recreated a mature heathland plant community. However, some changes in the plant community occurred which probably resulted from differences in soil drainage characteristics between the donor and recipient sites. M.caerulea and A.curtisii were not present, and P.serpyllifolia in 1991 only (0.3/m²). Seedlings of the other species were all present in good numbers: C.vulgaris (1990: 292; 1991: 484), E.tetralix (1990: 13.6; 1991: 16.4), E.cinerea (1990: 15.9; 1991: 34.2), U.minor(1990: 8.4; 1991: 8.3).

Conclusions: Heathland topsoil addition and turve transplants were both very successful in establishing heathland plant communities, at least over the short 2-year term of the study. Of the different sources of heathland plant propagules, harvested heather shoots however, were a renewable resource. The collection of heathland topsoil and turves involved the destruction of existing heathland but such material might be available during management to reinstate early successional habitats.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: