Providing evidence to improve practice

The journal, Conservation Evidence

Our online journal publishes research, monitoring results and case studies on the effects of conservation interventions. All papers include some monitoring of the effects of the intervention and are written by, or in partnership with, those who did the conservation work. It includes interventions such as habitat creation, habitat restoration, translocations, reintroductions, invasive species control, and education or integrated conservation development programmes, from anywhere around the world.

Watch a brief video on our journal here

A volume is created each year with peer-reviewed papers published throughout the year. We now accept Short Communications as well as standard papers.

Special issues contain new papers on a specific topic.

Virtual collections collate papers published in the journal on specific topics such as management of particular groups of species.

To search for papers on a specific topic within the journal select 'Advanced search' on the Home page, enter your keyword(s) and within the Source box type: "conservation evidence".

Habitat management



This virtual collection contains 49 papers on habitat management.

Coppicing is a commonly used management intervention to increase structural diversity in woodlands, but coppiced trees are vulnerable to browsing by deer. We investigated the effect of coppicing hazel stools at different heights on the survival of trees, and also the species richness of the ground flora. Plots were cut at experimental heights of 0.7 m and 0.8 m, with plots cut at 1.2 m and ground level as controls. All the stools cut at 1.2 m were alive five years after cutting. In the plots cut at 0.7 and 0.8 m, some shoots were eaten by deer but less than 10% of stools died. Less than 5% of stools in the plot cut at ground level survived.  After 7–8 years, coppicing at 0.7 m and 0.8 m supported a higher species richness of angiosperm ground flora than either of the control heights. We conclude that high-level coppicing offers a cost-effective opportunity to achieve a rotation frequency that increases tree survival and supports a diverse coppice-woodland angiosperm flora.

 

Land at Ranscombe Farm Reserve showed a build-up of biennial and perennial plants following a number years of conservation management for rare arable plants. The impacts of two different forms of cultivation were compared in order to understand how cultivation might be used to control this build-up, while maintaining the habitat for the rare arable plants for which the site is important. It was found that, in comparison with minimum tillage, ploughing produced lower overall plant cover but had no significant impact on the numbers of annual plant species, or on the number or population size of rare annual plant species. Plants considered to be problem species, such as creeping thistle Cirsium arvense and perennial sowthistle Sonchus arvensis, were not affected by the type of cultivation, but the abundance of these species did not appear to have a negative impact on those annual arable plants of conservation concern.

 

Changes in the vegetation of hay meadows under an agri-environment scheme in South Belgium
Piqueray J., Rouxhet S., Hendrickx S. & Mahy G. (2016), 13, 47-50

We monitored five-year changes in the vegetation of 31 hay meadows under an agri-environment scheme in Wallonia, Southern Belgium. Management included delayed mowing (in July) and fertilizer prohibition. It resulted in increasing cover of characteristic forbs (such as Leucanthemum vulgare, Lotus corniculatus, Centaurea jacea) and oligotrophic grasses (Avenula pubescens, Festuca rubra), while the competitive grasses, such as Holcus lanatus, Phleum pratense and Alopecurus pratensis, tended to decrease. We interpreted this as a vegetation shift from typical hay meadow to oligotrophic grasslands due to soil impoverishment following the current management. Both habitats are of conservation value. Despite these changes in the meadow plant communities, only one of the four criteria used by the Walloon administration to indicate hay meadow conservation status changed significantly over the six-year period. This was a decrease in the cover of species indicating high grazing intensity. The number and cover of characteristic plant species, and the cover of nitrophilous species, did not change significantly.

 

High conservation value grasslands, which are usually marginal and agriculturally poor, are often difficult to manage appropriately for biodiversity enhancement. A key management tool for this is conservation grazing, by which grazing intensity, timing and duration can be altered to suppress certain plant species, such as the more dominant grasses without impacting on other less competitive herbaceous ones. It has been suggested that the application of molasses to plant leaves could effectively encourage livestock to consume old and rank pasture grasses. This study assessed whether such an approach could be adapted to UK conservation grasslands, by using molasses to target grazing towards problem areas of dominant grass species. When Dexter cattle were exposed to areas of upright brome Bromopsis erecta and wood false-brome Brachypodium sylvaticum that had received a single application of molasses in the autumn period, no preference was shown for the treated plants. In the late winter period, however, cattle showed a significant preference for upright brome plants that had received two applications of molasses. Therefore, if consideration is given to the timing and frequency of molasses applied to target vegetation, it can be used as a conservation grazing management tool for some less palatable grasses.

 

We report the results of a nine year study of the effects of restoring low-intensity cattle grazing on the post-fire recovery of vegetation on the lowland valley mire and wet heath of Folly Bog, Surrey, UK. Four distinct vegetation communities were studied, with repeated recording of quadrats (n = 652) inside and outside grazing exclosures. Species richness increased across the valley mire, largely as a result of grazing-induced decreases in purple moor-grass Molinia caerulea and litter and increases in bare ground. Uncompetitive liverworts and waterlogging tolerant graminoids were particularly favoured. Purple moor-grass and litter removal also encouraged the spread of bog-mosses Sphagnum spp., although trampling in the wettest vegetation resulted in locally severe damage to the moss layer. On the firmer substrates of the wet heath, there were no such deleterious trampling impacts. Here, both bog-moss cover and species richness increased significantly, largely due to suppression of shade-producing heather Calluna vulgaris and litter, and the maintenance of bare ground. Our results reveal that the resumption of low intensity cattle grazing had many positive conservation benefits. However, site managers need to consider grazing on a site-by-site basis and retain flexibility to change stocking times and levels as conditions dictate. Other forms of management to supplement grazing will most likely continue to be required.

 

The effect of meadow management on plant species diversity was examined in a meadow in the west of England. In 2002 the meadow was assessed as species-poor. From 2002 to 2013 the meadow, along with 11 surrounding fields, was managed as a hay meadow, with grass being mown for hay in late July or early August each year and the aftermath then grazed by cattle. Vegetation surveys from 2002 and 2013 showed that the diversity of the meadow was significantly enhanced over the period of management, with ten additional meadow herb species becoming established by unaided colonisation. In consequence, a colourful, nectar-rich meadow has been created within the space of 11 years.  However, a number of species present on the farm that are more closely associated with old meadows have not yet colonised the field.

 

Restoration of a degraded dry forest using nurse trees at Dambulla, Sri Lanka
Medawatte W.W.M.B.A., Amarasinghe J., Iqbal M.C.M. & Ranwala S.M.W. (2014), 11, 16-19

We determined the effect of over-storey nurse tree cultivation on species composition in a naturally regenerated dry forest in a dry zone arboretum in Sri Lanka. The forest had previously been abandoned shifting cultivated land. One area was restored using nurse trees, one area was restored without nurse trees, and one area was left unmanaged as a control. Species dominance, richness and diversity of regenerated trees were assessed within random plots in the three treatment types. Regenerated tree species richness and diversity were greater in the restored land with nurse trees than in the restored land without nurse trees or in the control area. Dry forest tree species were dominant in the plots with nurse trees, while light-demanding and competitive pioneer scrubland species were dominant in the plots without nurse trees and the control area. We suggest that monospecific tree plantations that have been established for reforestation or agroforestry purposes could be used as nurse trees for dry forest restoration.

Fifteen-hundred Scots pine trees were ring-barked (as individuals, groups of five or 15) on Mar Lodge Estate, Scotland in order to create structural diversity and deadwood habitat in two plantations. A sample of 220 was monitored annually and compared with a control sample of 10 non ring-barked trees, to quantify structural changes as well as use by saproxylic invertebrates and woodpeckers. Eight years after ring-barking 26.1% (±6.13% S.E.) of the trees had snapped off and 2.4% (±1.37%) had fallen over completely; 48.0% (±12.5%) had lost 60-90% of their branches and 34.9% (±24.2%) had lost more than 50% of their bark. Additionally 98.5% (±0.92%) of the trees showed signs of wood boring invertebrates and 74.5% (±11.6%) were used by woodpeckers. Six species of beetle, four of which were saproxylic, and a single species of saproxylic fly were identified from fallen deadwood from the ring-barked trees. The control trees remained largely structurally unchanged and none were colonised by saproxylic invertebrates or woodpeckers. There were significant differences in structural change and use by woodpeckers between the two plantations but none in the occurrence of saproxylic invertebrates. Group size had no significant effect on colonisation, except for woodpeckers which used small groups of trees significantly more than larger groups. Ring-barking can provide an effective management tool to create structural diversity and deadwood habitat within a short period of time. However it is necessary to regularly repeat ring-barking in groups of different size in order to maximise structural variation and ensure niche diversity of such a dynamic substrate.

Concern over the decline in species-rich grassland in the UK has led to a focus on restoration. This study looks at the rate of natural colonisation of species into semi-improved grassland from adjacent unimproved species rich grassland over a 12 year period.  During this period the grassland had been managed traditionally with an annual hay cut followed by aftermath grazing and no input of fertilizer or farm yard manure.  During 2000 and 2012 vegetation surveys were carried out on two unimproved fields and two semi-improved fields.  These data were analysed for species-richness using two variables; Total species, and Wildlife Site Indicator species. A National Vegetation Classification survey was undertaken in 2012. Species-richness increased significantly in the semi-improved meadows during the study period.  These meadows now meet the criteria for Wildlife Site designation and the National Vegetation Classification community is shifting from MG6 to the target community MG5.

Deep Dale is situated within the carboniferous limestone area of the Peak District National Park.  The study site occupies an area of 36 hectares, representing the south-eastern slopes of the dale.  During the period from 1950 to 1996, the site was grazed by cattle, traditionally from the beginning of May each year.  Then, from 1997 to 2012, the grazing start date was delayed to the beginning of July in order to comply with the requirements of agri-environment schemes.  Repeat surveys indicate that this change in start date appears to have resulted in few pronounced changes to the vegetation.  Some areas of grassland on shallow soils (conforming to National Vegetation Community CG2d) have become more herb-rich with an increase in abundance of kidney vetch Anthyllis vulneraria, common milkwort Polygala vulgaris, devilsbit scabious Succisa pratensis and autumn gentian Gentianella amarella.  However, it appears that these changes are mainly associated with areas grazed preferentially (first) by livestock, whilst in an area of CG2d grazed later, fewer positive indicator species have shown an increase in their abundance and there are early signs of a decline in condition, including a decrease in the abundance of fairy flax Linum catharticum and an increase in the abundance of bryophytes.  Most significantly, areas of acid U4c grassland have shown a notable increase in the abundance of hawthorn Crataegus monogyna seedlings, and in the abundance of wavy hair-grass Deschampsia flexuosa.

 

At a site on Mar Lodge Estate, Scotland, a number of broadleaved trees were planted during the early 1990s. After fifteen years these trees were still barely higher than the tree tubes protecting them due to heavy browsing by deer. In 2004 a series of small exclosures were constructed around some of the trees using timber felled from a nearby conifer plantation. Fences were constructed with logs, which proved to be longer-lasting and sturdier than the woody debris used for protection elsewhere. The trees inside the exclosures are significantly higher than those which remain unfenced, and the ground vegetation has responded well. Deadwood fences have a number of benefits over traditional deer fencing: posing no threat to woodland grouse, having a lower visual impact in the landscape, and providing additional habitats for wildlife.

The Scottish Wildlife Trust established a mobile flock of sheep to manage its Sites of Special Scientific Interest (SSSI) designated grasslands in eastern Scotland. A programme of monitoring (quadrat sampling and 'Site Condition Monitoring') was established to record vegetation responses to sheep grazing over several years. The project has resulted in the sheep-grazed grasslands moving towards favourable condition in terms of target plant communities. It has highlighted the need to take a flexible and responsive approach to conservation grazing, and has demonstrated the usefulness and necessity of detailed monitoring in guiding changes to grazing regimes; the level of grazing 'fine-tuning' could not have been achieved, in this instance, using external graziers.

Trials were undertaken to assess the effectiveness of various treatments aimed at reinstating heathland vegetation at Hobbister RSPB Reserve (Orkney Islands) on a denuded area where no vascular plant growth had occurred since peat had been extracted commercially over 30 years previously. A management history of Hobbister was collated and information (derived from a literature search of restoration techniques) combined with observations of physical conditions at the site, was used to develop a list of possible impediments to heathland vegetation regeneration. Based upon these findings, eight sets of treatments were designed and applied to trial plots devoid of vegetation in June 2006. Plots were surveyed in August 2009. A combination of peat dust, heath mulch and geojute gave best results with 80% cover of vascular plants (including 70% by heather Calluna vulgaris). Although two grass-seed addition plots had higher cover values (91 and 86%) these were dominated by one of the sown species (red fescue Festuca rubra). Peat dust plus heath mulch addition also produced good cover (40%) of Calluna. Adding fertiliser did not assist in target heathland plant species re-colonisation. On the untreated control plot, vascular plant cover remained at zero.

 

A practical, cost-effective method of packing securely items destined for Round Island was required. Since 2001, re-sealable plastic barrels have been used for the transport of almost all materials onto the island. No unwanted 'stowaway' species have since been recorded when using this method of packing. The barrels are very versatile and are increasingly used in the transport of materials to other Mauritian offshore islets.

At a newly created wetland nature reserve in eastern England, a pair of water buffalo Bubalus bubalis was introduced with the aim of maintaining early successional habitats and creating a heterogeneous vegetation structure. The water buffalo grazed the required parts of the fen and reedbed, and created submerged tracks. These tracks may be used by fish to disperse into the reedbed and provide foraging areas for bitterns Botaurus stellaris.

 

Barriers made from willow Salix spp. bundles were installed along sections of the River Cam to protect the river banks from erosion. Subsequently, a more gently sloping river bank was created which was colonised by a range of riparian plants. These vegetated margins developed into an attractive wildlife habitat and are effectively protecting these river bank sections from further erosion.

Artificial macrophytes have been suggested as a means of improving water quality by providing zooplankton refugia. Plastic brushes provided a short term reduction in phosphates as they were absorbed by the periphyton growing on the brushes. They also provided a refuge for invertebrates. After two years, the brushes became colonised by sponges, which greatly reduced their long term usefulness as invertebrate habitat.

Native wetland species were planted within coir pallets to encourage plant colonisation along a shallow wetland margin in an attempt to improve water quality. Although some species initially grew well they were unable to withstand a period of hot weather and low water levels.

Removal of overhanging alder Alnus glutinosa and grey sallow Salix cinerea carr from the edge of an East Anglian broad led to a vigorous growth of riparian plants around the water's edge.

An island made of coir pallets supported by PVC floats was created with the objectives of producing an island of emergent vegetation and to cover a navigation hazard. Many of the planted species grew well and resulted in a reasonable cover of emergent vegetation. The island edges needed replanting where eroded by wave action.

Removal of alder Alnus glutinosa and grey sallow Salix cinerea carr from the edge of Hoveton Great Broad led to restoration of vegetation around the littoral margin; after removal of trees from some of the wetland edge emergent littoral plants showed vigorous growth.

A 15 ha saline lagoon was created in 2002 as part of a flood defence scheme at a site on the east coast of England. It has subsequently been used by a range of wintering and breeding waders and waterfowl.

Former arable land at Minsmere RSPB Reserve, eastern England, was treated with sulphur, herbicide was applied to control weeds, and seeds were sown in an attempt to create acid grassland. Soil pH was reduced and acid grassland target species dominated the vegetation three years after seeding.

Nine samples of soil were taken from former arable land at Minsmere RSPB Reserve, Suffolk, England. Sulphuric acid and iron was added. The pH of the soil was reduced, but the iron addition did not appear to have an affect on reducing the quantity of extractable phosphorus.

On former arable land at Minsmere RSPB Reserve, eastern England, sheep grazing was introduced with the objective of creating acid grassland. Seven years after the introduction of a grazing regime, the fields had lower cover and species-richness than the existing adjacent acid grassland.

An attempt was made to convert a former arable field to acid grassland. Elemental sulphur, bracken Pteridium aquilinum litter and heather Calluna vulgaris clippings were added and the area grazed with sheep. Over seven years the target acid grassland species cover increased considerably to 60.7%. Adjacent existing acid grassland had 85.6% cover of these species.

Sulphur and clippings of heather Calluna vulgaris and bell heather Erica cinerea were added to an area of former arable land with the objective of creating heathland. Nine years later these two species had both become well established.

Former cattle-grazed grassland and arable land were converted into wet grassland by raising the water level. Over the next five years the vegetation shifted towards plant communities characteristic of wet grassland.

At Berney Marshes RSPB Reserve, water levels were raised, foot drains were added, and grazing by sheep was introduced. The plant communities shifted towards communities' characteristic of lowland wet grassland. Breeding wading bird numbers increased in response to these habitat changes.

A faster and cheaper method of hedge-laying to benefit wildlife using a mechanical hedge laying technique has been developed. In comparison to traditionally laid hedges, those mechanically laid are broader at the base, thicker, taller, retain more deadwood and flower and fruit every year.

 

The effects of introducing cattle grazing to a saltmarsh on breeding redshank Tringa totanus were investigated. The density of breeding redshank did not noticeable change after introduction of grazing.

The effects of adding organic matter (barley straw) on aquatic invertebrate food supply for waterbirds in an artificial saline lagoon in eastern England was investigated. The addition of barley straw resulted in an increase in benthic invertebrate biomass the following year, but was not considered a successful long-term management strategy to boost invertebrate populations.

 

To restore an area of former heathland, soil nutrient levels were reduced by turf removal. Turf-stripping reduced dominance of wavy hair-grass Deschampsia flexuosa and promoted an increase in heather Calluna vulgaris.

In order to restore heathland, birch Betula trees were cleared, the humic layer was removed, and heather Calluna vulgaris seeds were spread out over the restoration area. Two years later there was a good growth of young heather plants with 5-10% cover seven years after the intial clearance, whilst some areas were dominated by wavy hair-grass Deschampsia flexuosa.

 

In order to restore heathland, Douglas fir Pseudotsuga menziesii and Scots pine Pinus sylvestris was removed, and thereafter three treatments were applied: i) no soil removal or seed addition, ii) seeding with heather Calluna vulgaris, iii) removal of humic layer and seeded with heather. There was little or no heather establishment in untreated areas; heather establishment and growth was good in one seeded area (15% cover) but not the other; where humus had been removed and seed added the area was dominated by common bent grass Agrostis capillaris (50% cover), Campylopus introflexus (20%) and juniper hair-moss Polytrichum juniperinum (13%); very little heather (<1% cover) had established.

Former arable fields were treated by removing the topsoil, adding sulphur and applying seed-rich cuttings of heather Calluna vulgaris, bell heather Erica cinerea and western gorse Ulex gallii using a muck spreader. After soil stripping an interesting plant community developed including some less frequent or rare species. Both heather species and western gorse germinated. Soil pH a year after the sulphur addition was 5.6 where applied at rate of 4t/ha and 4.2 where applied at 8t/ha.

 

An area dominated by purple moor grass Molinia caerulea was burnt, flail mowed, heather seed was added, and then grazed. Heather Calluna vulgaris seedlings were observed the next summer and grazing kept Molinia growth in check.

 

At a site in southern England in September 2004, about 4 ha of mature, dense, non-native, maritime pine Pinus pinaster was cleared using a shear-head timber processor. One year later in August 2005, the cleared area was predominantly covered in purple moor-grass Molinia caerulea (approx. 80% cover). The remaining 20% was either bare ground (approx. 15%) or mature heather Calluna vulgaris and gorse Ulex plants (less than 5% cover) present prior to clearance, but no heather seedlings were found. There was no evidence of any pine regeneration.

Approximately one third of a hectare of mature heather Calluna vulgaris was burnt in January 2000. Five years later, the burnt strip was revisited. It had fewer heather flowers and the sward was shorter (20-25 cm) than the surrounding vegetation (35-40 cm). Bristle bent grass Agrostis curtisii was more abundant (25% cover) than in adjacent unburned areas (5%).

 

A forage harvester was used to cut swathes of heathland vegetation at a site in southern England to increase habitat heterogeneity. Areas selected were predominantly dry heath or on the margins of humid heath and were cut to ground level. Six years later the cut areas were still clearly visible. In a humid heath area purple moor-grass Molinia caerulea had been greatly reduced, heather Calluna vulgaris had increased slightly, and open patches of bare ground (important for early successional plants) were present. On dry heath, heather cover was reduced substantially but the shorter and more open sward had allowed lichen communities to develop.

 

On a heathland in southern England, mature gorse Ulex europaeus was coppiced and the area fenced to prevent rabbit Oryctolagus cuniculus browsing in an attempt to create a varied gorse age structure. Twelve years later little gorse had regrown from the cut stumps and the cleared area had been invaded by bracken Pteridium aquilinum.

 

Approximately 0.5 ha of dense Scots pine Pinus sylvestris and maritime pine P.pinaster woodland was cleared using a shear-head timber processor with the objective of restoring heathland. Prior to clearance the area supported little heathland vegetation and was predominantly bare ground with some bracken Pteridium aquilinum, underneath the thick pine canopy. One year after clearance, there was deep litter layer (7-8 cm), little regeneration of heather Calluna vulgaris, and bracken was becoming dominant.

 

In January 2002, with the aim being to reinstate heathland vegetation, an area of 0.75 ha was cleared of Scots pine Pinus sylvestris and some maritime pine P.pinaster trees. Prior to clearance the area supported little heathland vegetation, predominantly comprising bare ground with some bracken Pteridium aquilinum. Three and a half years after clearance, the area supported a range of common generalist, non-heathland plant species, and was being invaded by silver birch Betula pendula saplings.

 

A number of pine Pinus trees were ring-barked using either single or double cuts, two trees also had their crown removed. One to two years later, all the ring-barked trees were still standing. The only two which had died were the ones with their crown removed. The remaining trees were all alive, but had some dead needles and needles showing signs of discolouration. There was no clear difference between trees cut using a single or a double cut.

 

Two adjacent mature Scots pine Pinus sylvestris trees were ring-barked using a chainsaw. Five years later, both trees had died. One tree had blown over just above the ring-bark cuts, leaving a jagged stump 1.25 m high; the other tree had lost its crown, resulting in 10 m of standing deadwood. Saproxylic invertebrates had colonised and great spotted woodpeckers Dendrocopus major had used the taller ring-barked stump for nesting. This management method proved to be a low-cost and easy way to produce standing deadwood.

During the winter of 1995-1996 on a heathland in southern England, 1.6 ha of invading silver birch Betula pendula (40% cover) and gorse Ulex europaeus were cleared using chainsaws and burnt on site. Ten years later, the cut birch and gorse stumps were still visible; the area was largely covered in dense bracken Pteridium aquilinum (average cover of 70% over 5 plots), the only gaps being the fire sites (predominantly bare) and a few patches of purple moor grass Molinia caerulea. Little heather Calluna vulgaris was present and there was considerable evidence of birch regeneration, suggesting that further management would be necessary to prevent invasion by scrub.

 

In 1991, 5.7 ha of Scots pine Pinus sylvestris scrub was removed from a heathland in southern England. Trees with diameters less than 25 cm were cut using bow saws and loppers; mature trees were left untouched. Fourteen years later, the area could still be identified by the presence of cut stumps; considerable pine regeneration (330 trees/ha) was apparent as there had been no ongoing management.

 

An area of about 13 ha of former open heathland in southern England was cleared of Scots pine Pinus sylvestris, maritime pine P.pinaster and birch Betula spp.; prior to management it contained 50-75% scrub and mature tree cover. Clearance was conducted using chainsaws; brash was burnt. Five years later, there was considerable evidence of pine regeneration (2,600 seedlings per ha). To maintain open heath, control of tree seedlings is required after tree clearance.

 

Dredging the silt from two old ponds in southern England (one in which the nationally endangered starfruit Damasonium alisma had recently been rediscovered) resulted in dramatic initial results, with many starfruit plants appearing. A subsequent rapid fall off in numbers suggests that seeds appear to germinate best on exposed sediment on drying margins and continuous low-level disturbance management is probably desirable.