Liley Et Al., 2006B)

Date: March 2010;

Version: FINAL

Recommended Citation: Liley D., Lake, S., Underhill-Day, J., Sharp, J., White, J. Hoskin, R. Cruickshanks, K. & Fearnley, H. (2010). Welsh Seasonality Habitat Vulnerability Review. Footprint Ecology / CCW.

Summary It is increasingly recognised that recreational access to the countryside has a wide range of benefits, such as positive effects on health and well-being, economic benefits and an enhanced understanding of and connection with the natural environment. There are also negative effects of access, however, as people’s presence in the countryside can impact on the nature conservation interest of sites. This report reviews these potential impacts to the Welsh countryside, and we go on to discuss how such impacts could be mapped across the entirety of Wales. Such a map (or series of maps) would provide a tool for policy makers, planners and access managers, highlighting areas of the countryside particularly sensitive to access and potentially guiding the location and provision of access infrastructure, housing etc.

We structure the review according to four main types of impacts: contamination, damage, fire and disturbance. Contamination includes impacts such as litter, nutrient enrichment and the spread of exotic species. Within the section on damage we consider harvesting and the impacts of footfall on vegetation and erosion of substrates. The fire section addresses the impacts of fire (accidental or arson) on animals, plant communities and the soil. Disturbance is typically the unintentional consequences of people’s presence, sometimes leading to animals avoiding particular areas and impacts on breeding success, survival etc. We review the effects of disturbance to mammals, birds, herptiles and invertebrates and also consider direct mortality, for example trampling of nests or deliberate killing of reptiles. Within each of these four sections we look across all types of access, but rather than single out particular activities, we highlight general mechanisms such as footfall or wear that result in the particular impact.

We include all terrestrial, freshwater and coastal habitats (but exclude coastal habitats that are restricted to the sub tidal and intertidal). We use broad habitat types as our focus, and for each impact we summarise the review according to habitat type, producing a matrix of 24 habitats and four seasons allowing direct comparison of different habitats throughout the year.

The final section of the report provides recommendations as to how the information in the literature review could be captured within a GIS to show the seasonal vulnerability of different areas of Wales. We recommend using a grid of 500m cells and a cumulative value calculated for each cell for each season. This score would be derived by summing individual scores (based largely on expert opinion) that capture the following:

 Soil types present with the grid cell (certain soil types are vulnerable to erosion)

 Habitats present within the cell (with comparative scores for each habitat finalised through expert opinion)

 Species present within or near the cell (a list of key species, both plant and animal would be finalised through expert opinion)

Further work is necessary to develop the actual maps and we identify the steps required for this to be completed.

Crynodeb Cydnabyddir fwyfwy bod mynediad i gefn gwlad at ddibenion hamddena’n esgor ar ystod eang o fanteision, megis effeithiau cadarnhaol ar iechyd a lles, manteision economaidd, dealltwriaeth well o’r amgylchedd naturiol, a chysylltiad cryfach â’r amgylchedd hwnnw. Fodd bynnag, gall mynediad i gefn gwlad esgor ar effeithiau negyddol hefyd, oherwydd gall presenoldeb pobl yno effeithio ar fudd safleoedd, o safbwynt cadwraeth natur. Mae’r adroddiad hwn yn adolygu’r effeithiau posibl hyn ar gefn gwlad Cymru, ac awn ymlaen i drafod sut y gellid mapio’r effeithiau ar draws Cymru gyfan. Byddai map o’r fath (neu gyfres o fapiau) yn adnodd i lunwyr polisïau, cynllunwyr a rheolwyr mynediad, a byddai’n tynnu sylw at y rhannau hynny o gefn gwlad sy’n arbennig o sensitif i fynediad, a gallai lywio gwaith lleoli a darparu isadeiledd mynediad, tai ac ati.

Rydym wedi trefnu’r adolygiad ar sail pedwar prif fath o effaith: halogi, difrod, tân ac aflonyddwch. Mae halogi’n cynnwys effeithiau megis sbwriel, cyfoethogi o ran maetholion, ac ymlediad rhywogaethau estron. Yn yr adran ar ddifrod, rydym yn ystyried cynaeafu a’r effeithiau a gaiff troedio ar lystyfiant ac erydiad isbriddoedd. Mae’r adran sy’n sôn am dân yn mynd i’r afael ag effeithiau tân (damweiniol neu fwriadol) ar anifeiliaid, cymunedau planhigion a’r pridd. Fel rheol, canlyniadau anfwriadol presenoldeb pobl yw aflonyddwch, sydd weithiau’n golygu bod anifeiliaid yn osgoi ardaloedd penodol, ac effeithiau ar lwyddiant o ran bridio, goroesi ac ati. Rydym yn adolygu effeithiau aflonyddwch ar famaliaid, adar, amffibiaid ac ymlusgiaid, ac infertebratau, ac yn ystyried marwolaeth uniongyrchol hefyd, er enghraifft, yn sgîl sathru ar nythod neu ladd ymlusgiaid yn fwriadol. Ym mhob un o’r pedair adran hyn, rydym yn edrych ar bob math o fynediad, ond yn hytrach na dewis gweithgareddau penodol rydym yn tynnu sylw at ffactorau cyffredinol megis troedio neu draul, sy’n arwain at yr effaith benodol dan sylw.

Rydym yn cynnwys pob cynefin tirol, cynefin dŵr croyw a chynefin arfordirol (ond yn eithrio cynefinoedd arfordirol sydd wedi’u cyfyngu i ardaloedd islanwol a rhynglanwol). Rydym yn defnyddio mathau cyffredinol o gynefinoedd fel ffocws, ac ar gyfer pob effaith rydym yn crynhoi’r adolygiad ar sail y math o gynefin, gan greu matrics o 24 cynefin a phedwar tymor er mwyn ein galluogi i gymharu gwahanol gynefinoedd yn uniongyrchol gydol y flwyddyn.

Mae adran olaf yr adroddiad yn rhoi argymhellion ynghylch sut y gellid cynnwys gwybodaeth yr adolygiad llenyddiaeth mewn System Gwybodaeth Ddaearyddol er mwyn dangos pa mor agored i niwed y mae gwahanol ardaloedd o Gymru yn ystod gwahanol dymhorau. Rydym yn argymell defnyddio grid o gelloedd 500m a gwerth cronnus wedi’i gyfrifo ar gyfer pob cell ar gyfer pob tymor. Byddai’r sgôr hon yn cael ei phennu drwy gyfrifo cyfanswm sgorau unigol (wedi’u seilio’n bennaf ar farn arbenigwyr) sy’n cynnwys y canlynol:

 Y mathau o bridd sy’n bresennol yng nghell y grid (mae rhai mathau o bridd yn fwy tueddol o erydu)

 Y cynefinoedd sy’n bresennol yn y gell (gyda sgorau cymharol ar gyfer pob cynefin wedi’u pennu drwy farn arbenigwyr)

 Y rhywogaethau sy’n bresennol yn y gell neu wrth ei hymyl (byddai rhestr o rywogaethau allweddol, sy’n cynnwys planhigion ac anifeiliaid, yn cael ei phennu drwy farn arbenigwyr).

Mae angen gwneud mwy o waith i ddatblygu’r mapiau, ac rydym yn nodi’r camau y mae angen eu cymryd i gyflawni hynny.

Contents

Summary ...... 2 Crynodeb ...... 3 Contents ...... 5 Acknowledgements ...... 7 Introduction...... 8 Contamination ...... 12 Introduction ...... 12 Litter ...... 12 Nutrient Enrichment ...... 13 ‘Alien’ plants ...... 16 Alien Animals ...... 16 Disease...... 17 Particular Habitat Issues and Seasonal Effects ...... 17 Damage ...... 21 Introduction ...... 21 Damage to substrate - mechanisms ...... 22 General effects of increased ground pressure on substrates ...... 22 Differences between recreation activities ...... 23 Natural factors interacting with trampling ...... 24 Intensity ...... 26 Damage to vegetation - mechanisms ...... 27 Impacts on individual plants ...... 27 Changes to vegetation communities ...... 28 Particular habitat issues...... 30 Broad-leaved woodland ...... 30 Semi-natural grasslands ...... 31 Lowland heathland ...... 33 Upland and Montane Habitats ...... 35 Bog...... 36 Fen and swamp ...... 37 Standing water/rivers ...... 38 Sand dunes ...... 39 Shingle ...... 40 Rocky shores ...... 40 Maritime cliffs ...... 41 Saltmarsh ...... 42 Saline lagoons ...... 43 Mud & sand flats ...... 43 Fire ...... 48 Introduction ...... 48 Soils ...... 52 Vegetation ...... 53 Birds ...... 57 Invertebrates ...... 62 Disturbance ...... 67 Introduction and definitions of disturbance ...... 67 Population consequences ...... 67 Mechanisms: the types of activity that can cause disturbance...... 70 Intensity of Disturbance ...... 71 Particular Species or Species Groups ...... 73

Mammals ...... 74 Birds ...... 77 Herptiles...... 78 Invertebrates ...... 78 Seasonality and temporal variation ...... 81 Particular habitat issues...... 82 Summaries by habitat ...... 86 Broadleaved, mixed and yew woodland ...... 86 Coniferous woodland ...... 86 Boundary and linear features ...... 87 Arable and horticulture ...... 87 Improved grassland ...... 87 Neutral grassland ...... 87 Calcareous grassland ...... 87 Acid grassland ...... 88 Bracken ...... 88 Dwarf shrub heath...... 88 Fen, marsh and swamp ...... 89 Bogs ...... 90 Standing open water and canals ...... 90 Rivers and streams ...... 91 Montane and upland habitats...... 91 Inland rock ...... 91 Built-up areas and gardens ...... 92 Saline lagoons ...... 92 Coastal vegetated shingle ...... 92 Rocky shores ...... 92 Maritime cliffs and slopes ...... 92 Saltmarsh ...... 93 Mudflats and sandy beaches ...... 93 Coastal sand dunes ...... 93 Mapping Seasonal Vulnerability ...... 95 Overview ...... 95 Use of a grid to summarise data ...... 97 Soil ...... 97 Topography ...... 98 Habitat ...... 98 Fire ...... 103 Plant Species ...... 103 Animal Species ...... 105 Overall Scores ...... 106 Contamination ...... 106 Damage ...... 106 Fire...... 106 Disturbance ...... 107 Appendix I: Welsh bird species potentially vulnerable to disturbance ...... 133 Species list ...... 134 Appendix II: Simplifying the Phase I to give habitat layers for the vulnerability map ...... 141 Appendix III: Example maps ...... 144

Acknowledgements This contract has been managed by Joe Roberts at CCW and we are very grateful to Joe for his support and enthusiasm throughout the project.

A number of other CCW staff have also provided specialist input, advice, support and expertise. We would like to acknowledge the following: Kathryn Birch, Paul Brazier, Adrian Fowles, Amanda Halstead, Barbara Jones, Richard Owens, Rachel Parry, Jonathan Rothwell, Jan Sherry, Jane Stevens, Sian Whitehead, Sue Williams.

Introduction This report provides a review of the nature conservation impacts of recreational access to the Welsh Countryside. In the past access and nature conservation have typically been viewed as opposing goals (Adams, 1996, Bathe, 2007), to the extent that nature reserves often restricted visitor numbers and access (e.g. through permits, fencing and restrictive routes). It is now increasingly recognised that access to the countryside is crucial to the long term success of nature conservation projects and has wider benefits such as increasing people’s awareness of the natural world and health benefits (Alessa et al., 2003, Bird, 2004, English Nature, 2002, Morris, 2003, Pretty et al., 2005).

There are however, circumstances where visitors can have a detrimental effect on the sites they visit, for example through disturbance or from trampling. There is a large body of scientific and grey literature addressing such impacts of access, and a number of general reviews on the effects of access are available (for example see Hockin et al., 1992, Kirby et al., 2004, Woodfield and Langston, 2004, Lowen et al., 2008, Nisbet, 2000, Penny Anderson Associates, 2001, Saunders et al., 2000, Liddle, 1997, Buckley, 2004). However the impacts and issues are complex and researchers tend to avoid making practical recommendations. It is therefore often difficult for conservation practitioners or policy makers to fully understand the implications of the research, let alone see a plan or project through appropriate assessment or understand when practical measures are necessary to avoid damage to the nature conservation interest of a site.

Human pressures on the natural environment are increasing, with an increasing population enjoying a wider range of recreational pursuits in the outdoors. The majority of adults (94%) resident in Wales had visited the outdoors at least once in the last 12 months, based on a very broad definition of places and activities (CCW and FC, 2009). Climate change may also exacerbate some issues, for example access patterns may change and the vulnerability of habitats to trampling and fire are both be related to weather conditions. The need for clear guidance on vulnerability is therefore critical to underpinning site management, policy and planning. The ability to compare between habitats and to map sensitive habitats and seasonal vulnerability will become increasingly important.

This report sets out such a review, presenting the evidence relating to the seasonal vulnerability of Welsh habitats. The review forms the foundation for further work to map habitats across Wales according to their vulnerability to impacts from recreational access. We consider all types of access and activities, but limit the review to terrestrial and coastal habitats, therefore largely excluding marine/inter-tidal habitats. In order to structure the review we consider four main types of impact, namely:

1. Contamination (e.g. litter, fouling, eutrophication). We include the spread of exotic species within this section 2. Damage (direct damage to vegetation through harvesting and wear and impacts to soils such as erosion) 3. Fire (both arson and accidental fire, resulting in direct mortality, habitat destruction, soil damage) 4. Disturbance (which can result in energetic consequences, direct mortality, increased stress, increased predation etc.).

Secondary impacts, for example where recreation creates complications for land management (such as implementation of grazing) with subsequent effects on the quality of sites, are beyond this review.

These four types of impact form the main chapters for the review. Within each chapter we summarise the impacts according to:

 The broad types of activities that have impacts, focusing on the particular mechanisms (such as footfall, wheels etc)  The intensity of use at which impacts may occur  The types of species and particularly vulnerable species impacted  Any variation in seasonality at which the impacts might occur

There is already an extensive body of reviews that address nature conservation issues and access (Bellan and Bellan-Santini, 2001, Carney and J., 1999, Hill et al., 1997, Hockin et al., 1992, Lowen et al., 2008, Nisbet, 2000a, Saunders et al., 2000, Sidaway and Ramblers' Association., 1990, Whitfield et al., 2008, Woodfield and Langston, 2004, Penny Anderson Associates, 2001a, Penny Anderson Associates, 2006, Underhill-Day, 2005, Liley et al., 2006b). Given the range of subject matter, involving different issues (such as trampling, erosion, disturbance), a wide range of activities (including access on foot, bicycle, boat etc) and the wide range of potential species and habitats, our approach was to use these reviews to form a basis for this report, supplemented with new material as appropriate. An initial database (in Endnote) was established, based on other reviews produced by Footprint Ecology (Lowen et al., 2008, Underhill-Day, 2005, Liley et al., 2006b). Each chapter was authored by a subject specialist, drawing from the shared reference database. Additional material was sourced, mainly using the following on-line bibliographic sources:

 Web of Science

 Google Scholar

 Science Direct

Search terms were tailored to the species/taxa/habitat and issue, and were used iteratively. Habitat or species terms were used alongside terms that included:

General: access, recreation, people, bicycle, foot, wheel, dog, horse, climbing, walking.

Contamination: contamination, alien plant, introduced, non-native, fouling, enrichment.

Damage: erosion, footfall, trampling, trample, pressure, friction, wheel, foot, disturbance, ground pressure.

Fire: fire, burn, burnt, incineration, heat, temperature.

Disturbance: stress, flight initiation, FID, flushing, flush, disturb, disturbance, crush, predation, approach.

At the end of each of the four chapters we generate a summary table highlighting the key issues, by habitat and season. For these tables we use the seventeen broad habitat types (terrestrial and freshwater) as defined by JNCC (Jackson, 2000), and a further seven coastal habitats (Saline lagoons, Coastal Vegetated Shingle, Rocky shores, Maritime cliffs and slopes, Saltmarsh, Mudflats and sandy beaches, and Coastal sand dunes). For each chapter (e.g. contamination), the table therefore consists of 24 habitats, with a column for each of 4 seasons. In these summary tables we also highlight the extent and robustness of the evidence, using colours to allow the reader to quickly pick out where there is a strong body of evidence, including a range of peer reviewed studies or where the evidence lack peer-review or is anecdotal. The below diagram illustrates the colours used. This colour coding is intended to provide a simple visual interpretation of the evidence base, rather than a rigorous systematic assessment of the quality of the literature. The tables were generated by each respective author and break a continuous scale into four broad categories:

 Dark green: strong body of evidence, for example there are at least five peer-reviewed studies published in national or international journals, and the different issues are comprehensively covered.

 Light green: reasonably strong body of evidence, at least some peer reviewed studies. Coverage possibly not comprehensive (perhaps not all species or aspects are well documented – e.g. there may be studies of the impacts of disturbance for only a small selection of key species associated with a habitat).

 Light grey: a reasonable body of evidence, but peer review and rigorous empirical studies largely lacking. Evidence base largely composed of research reports, grey literature and non peer-reviewed journals, but still a reasonable body of material.

 Dark grey: weak evidence, for example anecdotal reports and websites. Few references found.

Where an impact or effect on a particular habitat is known, however the seasons under which the habitat is vunerable is unknown or unconfirmed, then a single box will be presented covering all of the seasons. This is in comparison to those impacts or effects which a habitat is confirmed to be vunerable in all seasons and will be presented as four separate boxes, one under each season.

After the review we explore how the key impacts we have identified could be used within a GIS in order to derive sensitivity maps for the whole of Wales, highlighting areas potentially vulnerable (or not) to impacts from recreational access. In this later section of the report we do not actually produce such maps, but rather test whether such maps might be possible to produce and which

10 datasets might work to generate the information required. The long-term aim would be to produce a map for each impact type (i.e. contamination, damage, fire and disturbance) for each season (i.e. sixteen maps) as well as a single overall map which averages across all impacts and all seasons.

Contamination

Introduction Contamination of the natural environment as a result of access can take place by a variety of means through many different vectors. They can be loosely grouped into litter, nutrient enrichment, the spread of alien species and the spread of disease. These impacts can be associated with nearly all types of access, including by foot, vehicle, horse, boat, motorcycle or bicycle. Contamination of the marine environment, for example the use of anti-fouling paints on boats, is addressed by other authors (see Saunders et al. (2000) for review) and is not addressed here.

Litter Litter is a ubiquitous problem across any type of public space and can pose a large problem in the natural environment. It can range from large volumes of roadside fly tipping to a small number of discarded food wrappings. It can occur anywhere, regardless of habitat, although generally more prevalent in areas with greater public access. Despite the unpleasant visual impact it may have, there are other, potentially more damaging, effects that litter may pose in the natural environment.

Wildlife can become trapped inside, caught up in or injure themselves on discarded litter. While injuries can be relatively minor, it can also result in death, especially in the cases of entanglement or entrapment. This can include items such as bottles, small containers, broken glass or fishing equipment. For example it has been noted that although accidentally or deliberately discarded nylon line and netting have no impacts on the geomorphology of habitats, line and equipment discarded in intertidal areas can adversely affect wading birds by becoming tangled around their legs (Saunders et al., 2000). One study in 1965 noted that ten different species of British mammals were found trapped in discarded bottles (Morris and Harper, 1965), while a more recent study in North America found that 795, or 4%, of the discarded bottles found alongside roads contained either a small mammal, lizard or salamander (Benedict and Billeter, 2004). Alternatively, wildlife may ingest litter which can cause internal injury or infection and may also result in death. For example, the ingestion of discarded lead weights from fishing has also been shown to cause mortality in wildfowl (Locke et al., 1982).

Discarded food litter, be it deliberate or accidental, can pose a problem. In some cases animals can learn that people provide food and as a result may either endanger themselves by approaching people or vehicles or become dependent on it as their source of food. The consequences of both deliberate and accidental feeding of wildlife are well documented (see Green and Higginbottom (2001) and Higginbottom (2004) for review). Although no known studies have been conducted in Wales, there are many case studies of animals such as mice, raccoons, coatis, elephants, macaques, baboons and bears that have learnt to associate food with humans and have consequently changed their diet and behaviour accordingly (Buckley, 2004). In other cases food can attract rats and other scavengers, such as foxes and crows, which are in turn predators of other species. The association of crows (Neatherlin and Marzluff, 2004) and foxes (Harris and Rayner, 1986) with people is well

12 documented. In other circumstances, when broken down, food may also enrich the soil which can have local impacts to vegetation (see later section on fouling and enrichment).

Litter may also contain toxic substances which can pollute the soil, and may either find their way to and pollute a water course, kill or damage the immediate vegetation, or once assimilated into plants have damaging effects on those animals which consume them. Litter can also be non-biodegradable which can affect local vegetation by producing indefinite shading. There are no known studies on the impacts of litter to the natural environment by this mechanism.

Despite such concerns, litter and dumping of rubbish are rarely explicitly identified as a nature conservation issue, and there is cause for concern for few habitats – but see work on heathlands (Underhill-Day, 2005) and saline lagoons (Bamber et al., 1993).

The occurrence of litter in the natural environment can be deliberate, for example by people leaving behind the remains of their picnic, or accidental, for example blown out of overflowing refuse bins. The very sensitive issue of leaving memorials, such as plaques or mementos, to deceased family and friends in the natural environment can become a localised problem in a very small number of areas, such as mountain-tops or near significant landmarks. While they may be seen as beneficial to those placing them in the short term, they can pose the same risks as other litter to the natural environment and can be perceived as unsightly. This became the case at the summit of Ben Nevis where they now encourage visitors to leave memorials at a collective memorial at the base of the mountain (The Nevis Partnership, 2009). The environment agency also discourage the placing of memorials at sea, in streams and rivers or on riverbanks (Environment Agency, Undated).

While the information or number of studies that show litter as being a problem for wildlife is limited, there is a wealth of anecdotal evidence of its potential impacts. Nevertheless there is no evidence that any of the impacts of litter have any significant population consequences, and may only be of conservation concern to the rarest species. Similarly, very few studies attribute the impacts that litter can have to a particular habitat, however habitats that would be most vulnerable to the effects of litter are likely to be those supporting small to medium vertebrates such as mice, shrews, badgers, foxes, wading birds and wildfowl, and those that are susceptible to retaining broken down compounds, such as heavy clay soils. The seasonality of the impacts of litter will vary according to habitat and when the most vulnerable species are present or most active, however the persistence of litter within the natural environment may result in some impacts being present year-round.

Nutrient Enrichment The primary and most common route that access to the natural environment can result in nutrient enrichment is through fouling is by domesticated dogs. However human fouling can become a localised problem in remote areas or sites without toilet facilities, especially where camping takes place. Other methods can include discarded food waste and the scattering of human remains as ashes.

A number of reviews have addressed the impacts of dog fouling (Taylor et al., 2005, Taylor et al., 2006, Bull, 1998). Dogs will typically defecate within 10 minutes of a walk starting, and as a consequence most (but not all) deposition tends to occur within 400m of a site entrance (Taylor et

13 al., 2005). In addition most faeces are deposited close to the path, with a peak at approximately 1m from the path edge (Shaw et al., 1995). Similarly, dogs will typically urinate at the start of a walk, but they will also urinate at frequent intervals during the walk too. The total volume deposited on sites may be surprisingly large. At Burnham Beeches NNR over one year, Barnard (2003) estimated the total amounts of urine as 30,000 litres and 60 tonnes of faeces from dogs. Limited information on the chemical composition of dog faeces indicates that they are particularly rich in nitrogen and that modern dog food contain an excess of nutrients to improve flavour and any excess is excreted (see work cited in Taylor et al. (2006) and Taylor et al. (2005)).

Nutrient levels in soil (particularly nitrogen and phosphorous) are important factors determining plant species composition and on some habitats, for example heathland, the typical effect will be equivalent to applying a high level of fertilizer, resulting in a reduction in species richness and the presence of species typically associated with more improved habitats. Consequently a lush green strip is often evident alongside paths as nutrient enrichment can also lead to more vigorous growth and flowering (Taylor et al., 2006). The critical thresholds of the deposition of atmospheric nitrogen which is likely to cause a change in the vegetation in semi-natural grasslands range from 15 to 30 kg N ha-1 year-1, indicating low levels of deposition can have a significant impact (Taylor et al., 2005). Similarly, semi-natural grasslands generally have levels of available phosphorous below 8 mg kg-1 (Gilbert, 2000), while studies have shown that in areas of high dog use these levels are ten times higher (Bonner and Agnew, 1983).

One study on chalk grassland, a typically nutrient poor habitat, showed that in the first 50m alongside the path the typical chalk grassland flora was replaced by crested dog’s-tail and perennial ryegrass (Streeter, 1971). It also showed that although this change in flora did not correlate well with available soil nitrogen, it did correlate with soil phosphate, hypothesised to come from dog faeces. In another study on a heathland site frequently used by dog walkers, available soil nitrogen and phosphate followed the spatial distribution as dog faeces which peaked at 1m from the path and showed a conversion from a heathy to grassy sward (Shaw et al., 1995). The same was also true for sand dunes at a site receiving a large number of dogs, where nutrient enrichment was observed alongside paths, a zone of rye-grass was observed and the typical flora remaining tended to be more luxuriant and flower more profusely (Taylor et al., 2005, Milwain, 1984). Another example was found on saline lagoons where the use of adjoining beaches resulting in litter and eutrophication from faeces (Saunders et al. 2000). It must be noted however that trampling also has an impact on the floristic composition near paths and is therefore highly correlated with the occurrence of dog faeces, however it is thought that trampling exacerbates the problems that occur due to nutrient enrichment (Shaw et al., 1995).

Very little is known about the nutrient composition of dog urine and its impacts on habitats. It is however known that dog urine can scald vegetation and does provide some enrichment of soil nitrogen (Taylor et al., 2005). It is also known that urine does more damage on dry soils because the salts cannot disperse as easily. One study has shown that dog urine around the base of trees significantly alters the encrusting algal and lichen communities (Gilbert, 1989).

The persistence of dog faeces and nutrients in the soil will be subject to a number of factors, but primarily the soil type, soil water, weather and temperature. Dog faeces can take up to two months to break down, however if the weather is cold and dry this is likely to take longer, whereas if it is

14 warm and wet it is likely to take less time (Taylor et al., 2005). The persistence of these nutrients in the soil is strongly influenced by the soil type. In one study it was calculated that phosphorous derived from agricultural fertilisers persist between 15 and 20 years in sandy soils, while it was not uncommon for them to persist for 30 years or more in heavy clay soils (Gough and Marrs, 1990). It is therefore reasonable to assume that enrichment from fouling is likely to persist longer in habitats where the soil type is heavy and relatively imporous, compared to those which are light and free- draining. While the visual evidence of fouling only remains for a short time, the nutrients persist for far longer in the soil and therefore it can be considered that there is little seasonality in the vulnerability of a habitat to nutrient enrichment from fouling.

There is very little evidence to the extent of the problem of human fouling or the nutrient persistence in the natural environment. Problems, however, are likely to be highly localised. It is reasonable to assume that the visual and nutrient persistence of human faeces is similar to that of dogs, however there is no evidence to confirm this assumption.

Localised issues relating to human faeces tend to be restricted to mountainous areas used by hill walkers and mountain climbers, where day long or overnight trips are far away from toilet facilities. Fragile habitats on mountain plateaus are therefore susceptible to the effects of nutrient enrichment, and those who manage and care for some of our most famous and well climbed peaks within the UK are finding the issue of human faeces an increasing concern. For many years, mountaineering codes of best practice have promoted the burial of human faeces. However, the hole digging and burial can in itself cause damage to higher altitude flora, and the localised nutrient enrichment effect remains, albeit lower in the soil profile than if left on the surface. In recent years, the Cairngorms National Park Authority along with a number of partners has promoted the ‘Poo Project.’ The aim of the project is to raise awareness of the impacts of human faeces on the mountain environment, and to promote the use of biodegradable carry bags that can be put into specialist air tight pots until the bag can be properly disposed off once back down the mountain. Similar projects have been running for a number of years in the American wilderness areas, with the US Center for Outdoor Ethics leading the ‘Leave No Trace’ campaign, aspiring to educate those enjoying the American National Parks and wilderness areas to literally leave no trace of their visit, including when toileting in the wilderness. Both the Scottish Poo Project and the American Leave No Trace Campaign cite harm to biodiversity as one of the key reasons for the campaign, but research relating to the effects of human faeces on biodiversity are not quoted.

Bridle and Kirkpatrick (2005) investigated the rate of breakdown for buried toilet tissue and tampons in natural environments in Tasmania, Australia, and found that whilst buried toilet tissue had significantly broken down, tampons remained after the two years of their study. They therefore highlight the need to educate visitors to take sanitary items with them. The Bridle and Kirkpatrick study did not go on to discuss the implications of their findings for plant communities. However, assumptions can be made when considering the fact that cotton products generally contain a high degree of insecticides and pesticides. It can be concluded therefore that tampons potentially pose a risk to sensitive flora and soils. It is also worth noting that the chemical components of toilet paper, particularly bleached paper, are likely to have some impact on alpine floras, although again no studies specifically on this topic are known.

Another route by which enrichment through fouling can occur is from horses. There are no known studies into this form of enrichment, however it likely that some impacts do occur as a result of soil enrichment by horse faeces (Newsome et al., 2004) .

The scattering of cremated human remains in the natural environment is a sensitive subject and is unlikely to cause significant problems in the majority of habitats, however in popular locations, such as mountain-tops or beauty spots, or in very sensitive habitats, cumulatively it can have damaging effects. These effects include phosphate enrichment, changes to soil pH and stimulation of plant growth and as a result the Mountaineering Council of Scotland produce guidelines on the scattering of ashes on mountains (The Mountaineering Council of Scotland, Undated). The environment agency also produces guidelines for scattering ashes in or near bodies of water (Environment Agency, Undated).

‘Alien’ plants Globally, invasive non-native species are considered to be the most important threat to biodiversity after habitat loss (Mooney and Hobbs, 2000). In the UK a wide variety of introduced plants and animals has caused serious environmental problems (Manchester and Bullock, 2000) Alien plants compete with indigenous species for space, light, nutrients and water, and are often at an advantage because their new habitat is devoid of natural predators and controls (Newsome et al., 2002).

A study in North America has shown that paths provide a conduit for the movement of exotic species into a habitat (Benninger-Truax et al., 1992). This implies that users of paths transport seeds or plant material, via shoes, feet, hooves or wheels, through direct contact. This is certainly the case in one study which showed that plant seeds could be transported on the shoes of walkers for over 10 km (Wichmann et al., 2009). This could therefore also provide the mechanism by which diseases and pathogens can be introduced and spread within a habitat.

Horse faeces can be a source of alien plant species as they are likely to be grazing on plants outside a site, and then deposit the seed in a ready supply of nutrients, which aids germination (Siikamäki et al., 2006, Newsome et al., 2002). Although only documented in Australia, horses have also been shown to introduce and transport fungal pathogens (Newsome, 2003, Stammel and Kiehl, 2004).

Dog faeces are a well publicised medium for the transfer of disease to humans, however there is no evidence that these diseases pose a risk to wildlife. Since dog food is primarily domestic dog foods, the likelihood of the introduction of alien species through dog faeces may be very small, however this is still a potential route for the introduction of alien species and plant diseases and pathogens.

Alien Animals The impact of the introduction of alien or non-native fauna upon native species is wide ranging and can be catastrophic for our native species. The plight of the red squirrel against the introduced grey squirrel, and of the water vole against the predatory American Mink are two of the most well known examples of serious population declines as a result of the introduced alien species. Within the literature there are few examples of alien species being introduced as a consequence of recreation, but rather there are some species whose spread is aided by recreation, and mostly in aquatic environments. Edgar (2002) states that random fish introductions to ponds, i.e. those not associated with the purposeful stocking of fish for recreational or commercial fishing, dramatically increases in locations with public access. He describes the common practice of transferring fish

16 from garden ponds when they become too big or the pond is to be in filled, to ponds in publicly accessible places, with the person transferring the fish believing they have ‘done the right thing’ for the fish. Alien species within the aquatic environment cause significant reductions in native aquatic animals, and in some cases can irradiate native fauna completely.

Beebee (1997) found that the introduction of goldfish to two ponds within his long term study of ponds within the Sussex chalk downlands resulted in the extinction of the great crested newt populations. Both frog and toad tadpoles can also be consumed by introduced fish. Edgar (2002) notes that one of the most veracious predatory alien species is the terrapin, brought to ponds in publicly accessible places and released. Terrapins will feed on juvenile and adult amphibians, and can therefore quickly eradicate a pond of its newts, frogs and toads.

The zebra mussel Dreissena polymorpha originates from the Caspian and Black Seas, but has spread rapidly throughout the globe in freshwater habitats. It is commonly found in British waterways, and will readily colonise pipelines, causing significant problems when colonies grow and block pipes. Ecological implications for wildlife are most profound for our native uniloid mussels. Aldridge et al (2004) explain that zebra mussels will encrust the shells of our native species of mussel, which has the effect of starving our native species as they become unable to feed. This is of particular concern for the conservation of the native depressed river mussel Pseudanodonta complanata, recorded along the Welsh borders and listed in the UK Biodiversity Action Plan. Aldridge et al (2004) advises that waterways with significant boat use are more susceptible to invasions of zebra mussel due to the transportation of boats encrusted with the mussel between waterways.

A further instance of recreational boating aiding the spread of alien species and their disease is the spread of the crayfish plague, carried by the non-native signal crayfish. The spread of the disease is positively assisted by boat traffic moving through water (Josefsson and Andersson 2001).

Disease Pathogens may also be inadvertently transported by people or dogs. A dramatic and worrying example is the recent appearance of the plant pathogen Phytophthora ramorum, which causes sudden oak death (and also affects dwarf shrub communites). Studies in the U.S. have found that P. ramorum more commonly occurred in soils on heavily used tracks compared to soil from adjacent areas off trails. Human-induced dispersal occurred within already infected areas and into areas lacking local sources of inoculum (Hall Cushman and Meentemeyer, 2008). Advice from Defra indicates that Phytophthora pathogens can be spread on footwear within the UK.

Particular Habitat Issues and Seasonal Effects The impacts are summarised by habitat in Table 1. In many habitats it is difficult to specify seasonal vulnerability. For example much litter will persist in the environment for a long period, and therefore litter dropped will remain across seasons. Litter dropped in the winter (when there are fewer small mammals and reptiles present) is therefore likely to still be present in the spring and summer months. Similarly with nutrient enrichment, the issue from fouling is the steady build-up in nutrient levels accrued over an extended time period. The time scale for such impacts is over

17 multiple years rather than seasons. Nutrient enrichment is an issue that is particular to habitats/communities/species associated with soils with low nutrient levels, for example vegetated shingle (Randall, 2004), lowland heathland (Liley 2006; Shaw 1995).

Table 1: Summary of contamination by habitat. Colours reflect the evidence available for each issue (see introduction). Spring Summer Autumn Winter  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms (possibly for dormouse?), 1 Broadleaved, mixed and yew though no direct evidence. Impact all year as litter dropped in winter may persist all year round. woodland  Encrusting lichens and algae may be damaged by dog urine  Red Squirrel may become trapped in or poisoned by litter (no direct studies) 2 Coniferous woodland  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. Impact all year as litter dropped in winter may persist all year round.  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms (possibly for dormouse?), 3 Boundary and linear features though no direct evidence. Impact all year as litter dropped in winter may persist all year round.  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms (possibly for harvest 4 Arable and horticulture mouse?), though no direct evidence.  5 Improved grassland Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence.

 Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 6 Neutral grassland  Nutrient enrichment an issue, slightly less of an effect than other habitats as more mesotrophic?  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 7 Calcareous grassland  Nutrient enrichment an issue

 Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 8 Acid grassland  Nutrient enrichment an issue

 9 Bracken Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence.

 Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 10 Dwarf shrub heath  Nutrient enrichment a particular issue  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 11 Fen, marsh and swamp  Nutrient enrichment an issue, slightly less of an effect as dogs less likely to defecate in wet habitats?  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 12 Bogs

 Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms (possibly for water vole?), 13 Standing open water and though no direct evidence. Impact all year as litter dropped in winter may persist all year round. canals  Ingestion and entanglement of litter by wildfowl  Water vole may become Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms 14 Rivers and streams (possibly for water vole?), though no direct evidence. Impact all year as litter dropped in winter may persist all year round.  Ingestion and entanglement of litter by wildfowl  Small mammals (probably at very low densities) may become trapped in litter, unlikely to be a serious issue in conservation 15 Montane habitats terms, though no direct evidence.  Sensitive to nutrient enrichment 16 Inland rock No literature found/relevant issues  Small mammals may become trapped in litter, unlikely to be a serious issue in conservation terms, though no direct evidence. 17 Built-up areas and gardens Impact all year as litter dropped in winter may persist all year round.

Spring Summer Autumn Winter

1 Saline lagoons  Sensitive to nutrient enrichment

2 Coastal vegetated shingle  Sensitive to nutrient enrichment

3 Rocky shores No literature found/relevant issues

4 Maritime cliffs and slopes  Sensitive to nutrient enrichment  Ingestion and No literature found/relevant 5 Saltmarsh entanglement of litter issues  Ingestion and entanglement of litter by waterfowl etc by waterfowl etc  Ingestion and No literature found/relevant  Ingestion and entanglement of litter by waterfowl etc 6 Mudflats and sandy entanglement of litter issues beaches by waterfowl etc  Sand lizards may become trapped in litter (no direct studies), an issue all year round as litter can persist 7 Coastal sand dunes  Sensitive to nutrient enrichment

Damage

Introduction Recreational activities can lead to changes in soil characteristics and ultimately lead to erosion, the process by which sediment, rock, soil and other particles are removed. Although erosion brought about by recreational activities is small compared to natural factors it can none the less an important form of soil degradation (Holden et al., 2007). Changes to substrates can in turn lead to changes in the ecological communities they support.

This section considers the way in which substrates can be damaged by recreational activities, considers particular recreational activities which are most likely to contribute to damage and explores the potential impact of substrate loss and damage on different habitat types. We then look at the effects of substrate change on vegetation and invertebrates and the direct impacts of recreational activities on vegetation communities (direct impacts on fauna are addressed in the Disturbance (section 0).

Recreational activities most likely to cause damage to substrates and vegetation include walking, running, climbing, potholing, scrambling, scree-running, horse-riding, cycling, motorised vehicles, and tobogganing and ski-ing. Most recreation activity will result in a similar sequence of deterioration of unprotected soil and vegetation (see Figure 1) whether as a consequence of footfall, hooves or wheels. At lower levels of use, the main impact is on vegetation and is largely mechanical (Bayfield and Aitken, 1992, Liddle, 1997) while higher levels of use will also affect substrates. Light use may cause a slight decrease in vegetation cover, and a decline in the incidence of flowering. Bare ground may be colonised by trampling resistant species such as rye grass Loluim perenne and plantain Plantago major (Speight, 1973). Heavier ground pressure (referred to from here on as trampling) will result in greater loss of vegetation, with plants surviving between stones or in areas where wear is less. Significant erosion can be expected where the plant cover falls below 70% (Liddle, 1997), but erosion can commence before this level is reached (Kuss and Morgan, 1984). As loss of vegetation takes place, there is disruption and progressive loss of soil horizons by direct physical abrasion or loosening and indirectly by water and wind erosion. Important changes in soil structure and chemistry can result from compaction. Poor permeability to water can increase surface run-off, and reduced aeration can result in anaerobic conditions and poor root growth. The mechanisms for damage to substrates and vegetation are discussed below in separate sections. We then consider habitats and species for which trampling may be a particular issues. Information on indirect impacts to invertebrates is also included in this section. Change to substrates from contamination and fire are considered in Sections 1 and 0. Note that in many cases these causes of change will work in combination, for example, dog-fouling is most common around car parks, where trampling is also likely to be greatest.

Figure 1 Damage to soils and vegetation with increasing use (Bayfield and Aitken, 1992).

Damage to substrate - mechanisms

General effects of increased ground pressure on substrates Penny Anderson Associates (2001b) review the affects of recreational trampling, which are summarised below. It is important to note that the relationship between trampling intensity and impact on substrates tends to be curvilinear, with the greatest damage occurring with low use (Kuss et al., 1990). For example, compaction and erosion impacts are greatest at the early stages of use (Cole, 1986, Cole, 1987). Thereafter, the negative impacts of additional use slow considerably (Stankey and Manning, 1986).

Trampling (used here to mean exertion of ground pressure through any recreation activity) results in soil compaction. Compaction increases bulk density - soil pore spaces decline, reducing the amount of air and the infiltration rate for water. Larger soil aggregates can be destroyed. Clays change their form from a well structured, flocculated state when the particles are in a random arrangement, to a situation where the particles lie parallel, and have the highest bulk density. Surface pressure may be transmitted down through the soil profile. Liddle and Grieg-Smith (1975) found the most compacted level at c.15cm under a trampled zone on a sand dune.

The loss of soil pore space is greatest in the upper surface of the soil. Chappell et al. (1971) found a loss of 14% in the upper 2.5cm of a trampled chalk soil, but only a 2% loss in the 2.5- 5cm layer. A similar pattern has been found in sandy and sandy loam soils (Lutz, 1945). Air porosities of 19% or more are required for good plant growth, but heavily trampled soils frequently show levels below this. This can lead to oxygen deficiency and build-up of carbon dioxide.

Trampling also alters the amount of litter present. It is likely to initially increase the amount of litter as a result of damage to plants. On heather moorland, Bayfield and Brookes (1979) found increases in litter from 20% to 70% of the total biomass under heavy use as the vegetation was damaged but a reduction to 40% under severe use. However, even light trampling can cause a decline in litter. Duffey (1975) found an 81% decrease in litter

22 resulting from 10 tramples a month on grassland. Trampling can also cause litter to become incorporated into the soil.

In general, soils are likely to become wetter due to compaction and subsequent reduction of drainage. However, the effect of trampling on soil water content and its availability depends on the soil type. Clayey soils have greater potential to retain water by capillary forces due to the smaller pore spaces. However, retained water may not be available to plants if the suction pressure is too great.

Soil temperature may also change as a result of changes in air and water content, increases in bulk density, and loss of vegetation. Path surfaces may become warmer by day and colder by night. Liddle (1997) records a 9°C increase by day and 1oC decrease by night. The air above path surfaces may also affected – Liddle found air temperatures to be 6°C cooler than over adjacent vegetation by day and 2°C warmer at night. The extent of these differences will depend on the water content of the soil.

Soil chemistry is also affected by trampling. Liddle (1997) notes that soil pH appears to drift towards a pH of 5.5 – acidic soils tend to become more alkaline on heavily used paths, whereas pH drops with trampling on alkaline soils, although no satisfactory mechanism to explain this is provided. Nutrient levels in soil can also be altered by trampling, but different studies have shown opposing trends depending on soil type. For example, on chalk grassland, Chappell et al. (1971) found no significant differences in soil nutrient levels between trampled and non-trampled areas whereas (Leney, 1974) found higher levels of nitrogen and potassium on vegetation sand dunes and dune pasture compared with a bare pathway. Streeter (1971) found enhanced total nitrogen and phosphorus levels in heavily trampled areas of chalk grassland compared with areas of lower use, but very low levels in the bare paths where erosion was occurring. Changes in nutrients are likely to depend on the impact on soil micro-organisms as well as soil type. Liddle (1997) describes studies of the impact of trampling on bacteria: numbers of nitrifying bacteria (which raise soil fertility) were reduced by a factor of 10, whilst the numbers of denitrifying bacteria increased by the same proportion. Denitrification usually occurs in anaerobic conditions promoted by trampling. Soils on or adjacent to paths also become enriched through horse and dog dung – see section 0.

Differences between recreation activities Different recreational activities can have a significantly different impact. Table 2 summaries the impact of different activities. For example, walking is only likely to cause significant damage at high level of usage, while horse riding and vehicles may cause significant damage at very low frequency. The extent of impacts also vary, with cycling, rock-climbing, vehicles, and picnicking most likely to create localised rather than extensive damage, walking is more likely to cause extensive damage. In general, walking is likely to be less damaging that horse riding, cycling or motorised vehicles. For example, Weaver and Dale (1978) showed that showed that horses were substantially more damaging, and motor cycles slightly more damaging than hikers in grassland and woodland in the US Pacific Northwest. Thurston and

Reader (2001) suggest that mountain bikes cause the same amount of damage as hikers in deciduous woodland, although MacIntyre (1991) and Rees (1990) show that mountain bikes may cause slightly more damage than foot traffic depending on the type of habitat.

Table 2: Differing impacts of some recreation activities (after Bayfield and Aitken, 1992) Abrasion Impaction Rutting

Walking   

Cycling   

Horse riding   

Vehicles   

Rock climbing 

Ski-ing 

The impacts of skiing have been studied in several countries (Grabherr, 1985). The effects of off-the-road vehicles have been investigated in a variety of habitats, mainly in the USA. There have so far been few examinations of mountain biking (but see Wilson and Seney (1994). Horse riding has also been little investigated. A few studies compare different types of activity under the same site conditions e.g. (Dale and Weaver, 1974, Weaver and Dale, 1978, Weaver et al., 1979, McQuaid-Cook, 1978, Price, 1985, Summer, 1986).

Natural factors interacting with trampling Soils differ markedly in their trafficability (resistance to wear) depending on shear strength and stoniness (Liddle, 1997). Soils with a high percentage of coarse particles such as sand and stones are least affected (more energy is required to move the particles), while those high in clay or peat are most prone to erosion and degradation (Holden et al., 2007). For example, on the Pennine Way, the widest areas of path were found over peaty ground where the peat had eroded exposing the underlying mineral soil (Bayfield, 1987). Once a path cuts through the soil to a solid bedrock, erosion rates may be much reduced, but where the parent material is, for example, unconsolidated glacial or alluvial sediments (consisting of fine sands, silts or clays) then damage can be significant root (Root, 1972).

Trafficability is also affected by wet conditions, as water acts as a lubricant, allowing soil particles to rub together and soil compaction to occur (see Table 3). Waterlogged ground is much more susceptible to puddling and poaching than coarse-grained soils (Liddle, 1997). Work in the Cairngorms (Bayfield, 1987) show that irrespective of the level of use, path width increased with the surface wetness and roughness and also with the angle of slope along the path (see below).

Table 3: Resistance to wear of soil types in relation to wetness (after Bayfield and Aitken, 1992) Soil Rock Gravel Sand Loam Silt Clay Peat Wetness

low- Wet high high high moderate moderate high moderate

low- very low- Dry high moderate low low very low moderate low

Soil type will also influence the changes in water retention, air porosity, pH and nutrients as discussed above.

In general damage is generally likely to be greatest in wetter seasons. However, drought can also exacerbate erosion. Extreme drought can result in desiccation of the soil surface. This leads to cracking and disaggregation of the crust into small peds which are vulnerable to mechanical breakdown by trampling, in addition to wind erosion.

It is generally assumed that climate change will lead to a temperature increase of 0.8 – 2.0°C by 2050. This is likely to manifest itself in warmer and drier conditions in the late summer and early autumn coupled with more intensive summer and winter precipitation events (Simmons, 2003). This intensification of extremes suggests that climate change is likely to lead to increased erosion. Holden et al. (2007) identify erosion impacts of increased storms on organic substrate in the Welsh uplands, which could leave hillsides vulnerable to secondary erosion processes. However, warming will result in an extension of the growing season and a reduction of frost frequency (Holden and Adamson, 2002, Holden, 2001). Since vegetated soil is less susceptible to erosion than unvegetated soil, an extended growing season may allow vegetation to develop for longer and the lack of frosts mean seedling survival would be increased. The net effect could be progressive revegetation of existing bare soils (Holden et al., 2007).

Angle of slope is another significant factor influencing loss of vegetation and erosion, as shearing and compressive force resulting from trampling increase with slope angle. Bayfield (1987) found a greater increase in path width on steeper slopes up to 20° on Cairngorm. Weaver and Dale (1978) found that the 1000 passes resulting a 50% loss of plant cover in mountain pasture was reduced to 700 on a 15% slope whilst the 300 on a forest floor was reduced to just 50 on a 15° slope. A study in the Lake District (Coleman, 1981) tested the importance of a number of different variables relating to footpath erosion. Erosion (measured as path width, extent of bare ground or maximum depth) was found to increase with the square root of the slope angle and the square of the recreation pressure. These two variables also interacted with each other, while other factors, such as vegetation type, soil type and topographic position, also influenced the rate of erosion. A slope angle of 15–17° was suggested as a threshold to separate actively eroding from stable slopes.

It seems also that the effect of slope is complicated by behaviour. For example, Bayfield (1987) found that downhill walkers were found to take more steps and to have a greater

25 impact with each step than uphill walkers. They also tended to deviate more from the centre of the path than walkers going uphill, possibly because when walking uphill the field of view (and choice) is relatively restricted. The differences increased with slope steepness. In contract to walking, Weaver and Dale (1978) noted that motorcycles had least impact on downhill slopes, due to exerting lesser downward forces than walkers or horses. Similarly mountain bikes will cause less damage downhill, assuming that the wheels continue to turn rather than skidding which can loosen track surfaces, move material downslope, and promote the development of ruts which channel water-flow (Cessford, 1995).

Intensity The threshold beyond which trampling has an effect is likely to vary between habitat types. Cole (1987) showed that for montane forest communities in the US, the relationship between trampling effects and vegetation loss, species loss and decreased penetrability was curvilinear. However, once mineral soil had been exposed, the relationship between trampling and mineral soil exposure was linear. The effects of trampling on vulnerable habitats is explored further in Particular habitat issues below.

A curvilinear relationship between the width of the bare part of a path and the number of users, with an initially rapid expansion and subsequent more steady increase in width, is also noted by Liddle (1997). He cites examples from the northern Rocky Mountains, where Weaver and Dale (1978) found that a doubling of the trail width resulted from every ten-fold increase in the number of walkers while Satchell and Marren (1976) found chalk grassland path widths increased 4.4 times for every ten-fold increase in passages. Lance, Baugh and Love (1989) suggest, based on work in the Cairngorms, that once a path has begun to erode, this is likely to continue even if the traffic along it reduces.

In practice, trampling intensity depends on behaviour. For example, some substrate and habitat types allow walkers to spread out, distributing the impact over a wider area, whereas other focus the impact on a narrow path. Bayfield (1973) found that the roughness of adjacent ground affected path width in the Cairngorms, with paths being where they passed through the roughest terrain (provided the path surface remained of the same quality). Path quality itself can also determine off-path use, for example 92% of walkers on the Pennine Way stayed on a path with a well-built new surface (Pearce-Higgins and Yalden, 1997) whereas 68% stayed on the path where deep bare peat was exposed and trampled (Yalden and Yalden, 1988). In general, more people will remain on a path where the adjacent substrate or vegetation provides a less comfortable option (see below also). Penny Anderson Associates (2001b) suggest that additionally off-path usage depends on whether desire lines coincide with paths and the attractiveness of the adjacent environment. Keirle and Stephens (2004) found that the majority of walkers in open countryside such as at Cwm Idwal remain on footpaths. The location of those that stray from paths is likely to be site specific and dependent on specific features, for example, a view point or “linear handrails” such as streams or unmaintained trackways. Where access does result in high path densities this can have particular impacts, for example causing fragmentation within a site.

Damage to vegetation - mechanisms

Impacts on individual plants The susceptibility of plants to trampling caused by recreation activities is influenced by a number of factors, which are reviewed by Kuss (1986). The genetic constitution of the plant, its adaptive flexibility, and anatomical differences inherent to growth habit and morphology determine its susceptibility. Conditions such as too much or too little moisture, nitrogen or oxygen starvation, and frosts can cause physiological injury and therefore increase a plant’s sensitivity to impacts. Environmental factors that may increase or lessen plant sensitivities to impacts are soil moisture levels, canopy density, elevation, aspect, microclimate, soil drainage, texture, fertility and productivity. Season, frequency and extent and type of activity will also affect impacts, which are further complicated by inter- and intraspecific competition, antagonism, and commensalism may influence differences in the sensitivity of plant communities to impacts.

Grime’s (1979) triangular model of three basic plant strategies to cope with different degrees of stress and disturbance is a useful framework to consider the likely effects of trampling on plant species. Stress tolerant plants have an inherently low capacity for growth and low competitive ability. They are often evergreen, long-lived, and flower and set seed irregularly. In contrast, competitive plants grow rapidly in productive soils in summer, typically die back seasonally, but are not tolerant of stress or disturbance. The third group, ruderals, are often short-lived, but are opportunistic colonisers after disturbance, usually of bare ground.

The relationship between productivity and resistance to wear is explored by Liddle (1997), who models the relationship between primary productivity and the number of passes required to reduce different vegetation types by 50%. In general, lower productivity swards (containing more stress tolerant plants with an inherently low capacity for growth) are less able to withstand trampling. Similarly Kellomaki and Saastamoinen (1975) correlated fertility and resistance to wear. Habitats such as chalk and acid grasslands, old hay meadows, moor and heaths are on infertile soils where productivity is low, and are likely to be vulnerable to trampling. However, individual species within these communities may be ruderals which will tolerate, even prefer, a degree of trampling. For example species such as slender centuary Cicendia filiformis are often found along damp tracks across heathland (Stewart et al., 1994). Such species may benefit from a reduction in competition from more vigorous species, from the creation of regeneration niches, or, like pale butterwort Pinguicula lusitanica, from the exposure of buried seed banks (Lake, 2002).

In general terms, plants which are more tolerant of trampling are grasses or rosette plants where the bud or apex are protected from direct damage, and which can regenerate rapidly after damage. For example, daisy Bellis perennis, greater plantain Plantago major, bucks- horn plantain Plantago coronopus, annual meadow-grass Poa annua, perennial rye-grass Lolium perenne and crested dogs-tail Cynosaurus cristatus are often found in strips in trampled zones beside bare paths if the soils are suitable. Some plants, such as annual

27 meadow-grass and greater plantain develop a diminutive genotype as a response to trampling or tight mowing (Warwick and Briggs, 1978, Warwick and Briggs, 1979).

More sensitive species are likely to be low growing plants with woody stems and slow recovery rates. Plants with tall herbaceous stems, particularly if they are over 30cm tall soft, hollow, or brittle and solid, and without basal leaves, are likely to be susceptible. Older, taller, woody stems on plants like heather tend to be more susceptible than younger ones. Other characteristics include thin leaves, a woody stem base, rigid leaves, and long leaf petioles (Liddle, 1997).

Most damage to vegetation is a result of mechanical damage to above ground biomass (Liddle, 1997). Trampling reduces leaf area, leaf thickness, width, length and number per plant, although species are differentially affected. Flowering and fruiting can be inhibited in susceptible species, and bulk and other underground storage organ weights in species such as bluebell Hyacinthoides non-scriptus can be reduced (Blackman and Rutter, 1950). Food reserves in the plants are also reduced and may take some time to replace. Although impacts on roots can be severe, compaction levels low enough to prevent root extension completely have not been recorded (Liddle, 1997). With a reduced leaf area, carbohydrates are used to regenerate leaf structure rather than being used for root growth, and plants may become shallow rooted. This can be critical under drought conditions. Through reducing water infiltration, compaction can also reduce the wetted depth of soil – the zone in which most plant roots grow. In compacted wet or waterlogged soils, an aerobic layer develops and plants not adapted to these conditions. The impact of reduced oxygen availability through loss of soil pore space is not clear, and is likely to vary between species (some species such as cottongrass Eriophorum spp. can take oxygen efficiently from the atmosphere to the roots). Similarly, the physiological effects of nutrient availability in compacted soils has not been researched widely.

Changes to vegetation communities The relationship between trampling intensity is and vegetation loss maybe curvilinear (e.g (Bell and Bliss, 1973, Hylgaard and Liddle, 1981, Cole, 1987, Kuss and Hall, 1991) or linear (Cole, 1995), with damaging only occurring once a certain threshold intensity has been reached. The relationship between trampling intensity and species diversity may also be curvilinear, reflecting an initial sharp decline in the numbers of more sensitive plants, followed by slower reductions in the more resilient species until bare ground develops.

The threshold at which point the response changes is likely to depend upon the site characteristics (such as wetness, slope, climate) and season in addition to the resilience of the plants species in the community and the relationship between them. For example, plants are generally less toleration of trampling when the ground is wet - Rees (1990) found substantially heavier damage in birch woodland and grassland when the ground was wet. In contrast Bayfield et al (1981) found the opposite trend in lichen rich swards, with much higher damage during dry ground conditions, when lichens are brittle. The high number of variables means it is hard to predict the quantitative threshold for any particular site,

28 although some generalisations can be made. For example, species diversity will generally be higher in mesotrophic environments where biomass is reduced regularly, and in more nutrient poor environments where biomass is removed occasionally (Bakker, 1998).

In their review of a number of studies of trampling, Penny Anderson Associates (2001b) point out that while these can be used to provide an indication of the vulnerability of species and habitats, they can only be applied to provide an assessment of likely impacts at other sites. In addition, the cumulative effects of underlying long-term trampling may be significant – for example (de Gouverain, 1996) found that past trampling impacts on soils may have long-term effects on the successional development of the plant community in subalpine woodland in US. However, many studies examining current or recent experimental trampling pressure do not consider this.

Different vegetation communities also exhibit different degrees of wear according to the extent to which the vegetation structure limits or modifies recreational movement (Bayfield and Aitken, 1992). For example, short grassland generally has high resistance to wear, but low impedance of movement, so trampling is likely to produce widespread scuffing. Bog and fen have poor wear resistance but high impedance to movement, with the result that there tend to be few but heavily worn paths, often meandering between tussocks of vegetation. Habitat type also tends to define the type of activity that takes place. For example, given the wet peaty nature of the substrate, bogs are potentially vulnerable to vehicular activity, cycling and horse-riding, but these activities are unlikely to occur.

Table 4: Trafficability and impedence of recreation movement according to broad habitat type. (after Bayfield and Aitken, 1992) Impedance of Broad habitat Trafficability recreational Common patterns of wear type movement

Short grassland high low Extensive scuffing, few bare paths

Tall grassland moderate low-moderate Extensive braiding

Few heavily worn routes, some local Woodland low - moderate high braiding

Bog and fen low high Few, heavily worn routes

Patchy heavy wear on cliff top and Cliff low high bottom and on ledges

Few heavily worn routes and some Sand dune moderate moderate- high extensive narrow braiding

Heathland low - moderate moderate- high Few, densely worn routes

Few, heavily worn trails and some Mountain low - high low-high extensive braiding and scuffing

Liddle (1997) notes that vegetation is generally more resilient to wear in the growing season however, low productivity vegetation may take hundreds of years to recover from wear - the time for full recovery is often underestimated.

Particular habitat issues

Broad-leaved woodland Plants adapted to shade often have relatively large leaves and thin cell walls, making them particularly vulnerable to trampling damage e.g. bluebell and enchanter’s nightshade (Cole, 1987, Cole, 1978). Initial introduction of trampling effects are particularly significant with 50% ground flora cover being lost with as little as 40 passes in one year (Cole, 1987). Rates of damage to British woodland ground flora (homogeneous stands of bluebell Hyacinthoides non-scripta, bracken Pteridium aquilinum and bramble Rubus fruticosusi) were most rapid at the initial stages of trampling (Littlemore and Barker, 2001). The tolerance to trampling was more a function of the ability to recover, rather than resist damage. In the example cited, the susceptibility of bluebell was some 12 – 15 times more than for bracken or bramble.

A dense canopy favours more typical woodland ground flora with larger leaves, compared with grasses in more open woodland, and this is more susceptible. Broad-leaved plants in woodland tend to disappear first, even under low trampling damage (Kellomaki, 1973). Wood pasture, with a higher proportion of grasses than typical woodland ground flora, is thus likely to be more resilient. Grime (1979), found more stress-tolerant species in the woodland ground flora of ancient woodlands, compared with secondary woodland, and stress tolerators with low productivity are more susceptible to trampling (Liddle, 1997). Ancient woodlands generally are more susceptible to damage from trampling than secondary woods. Trampling later in the season is less damaging than earlier when leaf growth and food storage are at maximum levels. Damage from trampling was more significant than flower picking for bluebells (Peace and Gilmour, 1949) over an 8-year study; and this was due to leaf damage and significant reduction in bulb weight (Blackman and Rutter, 1950). All above-ground parts of plants e.g. bluebell and dog’s mercury Mecurialis perennis, were lost in woodland subject to heavy use for paintball games (Thomas, 1991).

Wet woodland communities are particularly sensitive to trampling (Webster and Adams, 1989), with effects on marsh fern Thelypteris palustris, opposite-leaved golden saxifrage Chrysosplenium oppositifolium, marsh marigold Caltha palustris and sedges Carex spp. all noted under heavy woodland activity.

Trampling can result in damage to tree root growth, possibly associated with an impact on fungal mycelia (Dunn, 1984), while Speight (1973) considered that tree death could result from root exposure and damage. This could affect any trees near access routes but might be

30 especially the case for veteran trees, often with associated lichen interest, in woodland or wood pasture, that may be preferentially approached (Read, 2000).

A shift in species composition is also possible. Most of the species that grow well on heavily compacted forest soil are non-forest species tolerant of trampling (Godefroid and Koedam, 2004b). Trampling decreases plant cover and changes the ground flora species composition, altering it locally on paths and providing opportunities for new species to establish in previously unbroken forest vegetation. Species such as grasses better adapted to sunny, warm and dry conditions replaced more sensitive forest species such as dwarf shrubs and mosses. The effect of path-trampling on species composition and density extended for 8 m off-path. Trampling may change the micro-climate on the path, which may have a negative effect on mosses away from the path and off-path (Hamberg et al., 2008). The work in Belgian forests produced similar results, with forest paths being demonstrated to have a significant effect on the surrounding plant assemblages (Godefroid and Koedam, 2004a). The presence of a path results in an increase in the number of ruderal species, disturbance indicators, nitrogen-demanding species and indicators of basic conditions. Eutrophication and pH increases, as inferred from the plant composition, are perceptible up to at least 10 m from the path.

Studies in Belgian forests (Godefroid et al., 2003) show that ground flora takes a long time to recover both terms of density and species composition subsequent to the cessation of trampling. A single intensive event of orienteering caused damage to bryophyte and lichen communities under Scots pine in New Forest that had not recovered within a year (Thomas et al., 1994). Bilberry Vaccinium myrtillus, cowberry V. vitae-idaea and tufted hairgrass Dechampsia caespitosa were rather more resilient under conditions of 50 passages/day for 10 days in a Russian pine forest (Rogova, 1976). These impacts could apply to deciduous woodland, or in the case of lower plants, other open habitats.

Semi-natural grasslands Grassland has been found more resistant to trampling than forest vegetation. For example, 10,000 visitors/year were required for the ground to become bare in chalk grassland and 20,000/year were required before tracks of 1.5 m wide became established (Goldsmith, 1983). Harrison (1981) found a 50% reduction in plant cover on various grassland and heathland plots after 400 passes, while 1,600 passes caused 90% reduction.

In general, semi-natural grasslands of high nature conservation value are naturally low in nutrients, favouring low-productivity species of low competitive ability - the species generally most vulnerable to trampling damage. On the other hand, unlike woodland, grasslands tend to have a high proportion of grasses and rosette-forming herbs that are morphologically adapted to grazing and have more resilience to trampling, at least at light levels. Most of the species that can spread into trampled zones are general grassland components and not indicators of higher quality semi-natural habitat. On chalk grassland in southern England (Streeter, 1971, Chappell et al., 1971) progressive disappearance of more sensitive species such as squinancywort Asperula cynanchica, rough hawkbit Leontodon hispidus, salad burnet Sanguisorba minor and wild thyme Thymus drucei was noted under moderate to high levels of trampling. Their replacement by the more trample-tolerant

31 species such as ryegrass Lolium perenne was noted, in Streeter’s work, in a zone extending up to some 50m from the bare path and this was correlated with an increase in available phosphorus in this zone, possibly from the effects of dog faeces and urine being particularly associated with the path edges.

Traditionally, many grasslands of conservation value will be managed by grazing and/or cutting and will already be to some extent adapted to conditions comparable with a degree of trampling. The effects of trampling on unmanaged grassland can thus be expected to be more noticeable than on a short, grazed sward. In a study of tussocky fescue Festuca grassland in USA (Cole, 1987), it was shown that with 400 passes the vegetation cover was reduced to 50%; and 50% of the species originally in the sward were lost after 600 passes, with disproportionately most lost in the earlier impact. Neutral grassland in Russia showed similar rapid decline in species cover (Rogova, 1976), with only moderate trampling pressure; and the frequency of trampling also affected the ability of some more resistant species to show signs of recovery between treatments.

A comparison of calcareous and mesotrophic grasslands on Salisbury Plain (Hirst et al., 2005) demonstrated that calcareous grasslands were less resilient following disturbance than the mesotrophic grasslands, with slower colonization of bare ground and target species re- assembly. (This was historic military disturbance rather than trampling, but the relative response and time for recovery are of relevance). Mesotrophic grasslands typically took 30– 40 years to re-establish following damage, whereas calcareous grasslands took at least 50 years. Moreover, even after long time periods, there remained subtle but significant differences between the vegetation composition of the damaged and undamaged swards. Perennial rosette-forming herbs persisted at higher frequencies in swards disturbed 50 years ago than in undisturbed swards (Hirst et al., 2005).

Research in Canada has confirmed that grassland species diversity changes with levels of trampling disturbance (Vujnovic et al., 2002). Lower vascular plant, moss and lichen species diversity was found in undisturbed and lightly grazed as well as in highly disturbed plots. Intermediate levels of disturbance gave the highest species diversity. The species richness and diversity of exotic plant species also showed a significant positive relationship with the magnitude of the disturbance (Vujnovic et al., 2002). Trampling damage by horses was also implicated in a study in Eastern Europe, with an increase in recreational horse riding causing a shift from meadow communities of high conservation value being changed to horse pasture (Young et al., 2004).

The diversity and abundance of invertebrates also has been shown to be negatively affected by treatment of grassland, including the effects of trampling (Atkinson et al., 2004). A study was conducted by Duffey (1975) on the effect of trampling on the invertebrate fauna of grass litter in a 7-year old grass ley under very low trampling intensities (5 or 10 tramples/month). This showed that the invertebrate fauna of grassland litter is affected by trampling pressures which were much lower than those needed to induce changes in the structure and species frequencies of the living plants. The species changes seemed to have been caused by changes in the structure of the litter habitat, (bearing in mind that the

32 grassland was a young ley and not species-rich), in particular, the smaller volume, smaller air spaces, fragmentation of leaves and stems, and the creation of a mud-litter mixture.

Despite the general trend for invertebrates to be negatively affected by trampling damage, there is a range of grassland species that are associated with bare ground and open vegetation which the effects of access may maintain. These include a number of bees and wasps associated with chalk downland (Alexander et al., 2005d) such as nomad bee Nomada signata, scabious bee Andrena hattorfiana and another mining bee Andrena tibialis. The ground bug Heterogaster artemisiae is associated with thyme Thymus drucei on coastal dunes, cliff tops and chalk grassland. This ground bug seems to show a preference for areas with broken or partly bare ground or scree, or the edges of tracks, where there are large clumps of thyme over bare ground or stones (Kirby, 1992). Thyme growing in closed turf appears to be unsuitable, and light disturbance from public access and recreation may maintain suitable conditions for this species. There are similar habitat requirements for a number of species of lowland acid grassland; indeed, there is much species crossover between different grassland habitat types. For some species such as certain mining bees and solitary wasps the key factor appears to be light trampling, as from walkers, but not excessive disturbance from activities such as horse riding and motorcycling (Lowen et al., 2008).

Lowland heathland Heather-dominated vegetation is very susceptible to trampling damage, though there may be some differences related to individual species response and soil conditions. In summer and winter trials on undamaged lowland heathland in England (Harrison, 1981), it was shown that 400 passes in the first summer of the experiment, caused heather cover to fall to about 50%, and by 800 passes it was less than 10%. The vegetation failed to recover in the period following the experimental trampling, after winter only, summer only, or all season trampling.

Seasonal and habitat response was tested in trials on heathland in Brittany (Gallet and Roze, 2001) and though there were some differences, in all cases trampling led to a great decrease in vegetation cover, with the vegetation cover varying between 0 and 50% under 750 passes. Dry heathland was more resistant than mesophilous (humid) heath and significantly so with winter trampling, but both heath types were equally vulnerable in wet conditions. Gorse was more resilient than heathers; and younger dwarf shrubs were less vulnerable than older plants.

Heather is also more susceptible to trampling damage than purple moor-grass Molinia caerulea (Lake et al., 2001). In Belgium, Roovers et al. (2004) found that dry heath with a high proportion of grasses – purple moor-grass and wavy hair-grass Deschampsia flexuosa - as well as dwarf shrubs, was less sensitive to trampling.

Though trampling can damage the dwarf shrub community of heathland, there are some aspects of the habitat that need the canopy to be broken, even to the extent of bare ground being sustained. Bare ground and early successional habitats are a very important

33 component of the heathland ecosystem, important for a suite of plants, invertebrates and reptiles (Byfield and Pearman, 1996, Lake and Underhill-Day, 1999, Moulton and Corbett, 1999, Key, 2000, Kirby, 2001). Typically small, low-growing herbs with low competitive capacity require these open conditions and lack of suppression by a taller canopy. Some may be ruderals or annuals that can only survive in such conditions. Some kind of physical disturbance is usually required to create these bare ground habitats, and hence a certain level of physical disturbance, including erosion resulting from trampling, can be beneficial. However, the level of disturbance required is difficult to define and is likely to vary between sites (Lake et al., 2001). There are likely to be optimum levels of use that maintain the bare ground habitats but do not continually disturb the substrate. Such levels of use have never been quantified, nor is it known whether sporadic use is likely to be better at maintaining bare ground habitats than low level, continuous use. It has been noted that marsh clubmoss Lycopodiella inundata tends to colonise heavily disturbed sites once the pressures are reduced or relaxed 1 . Heavy continuous disturbance was thought to be responsible for the failure of the plant to survive on a footpath at a site in Surrey. Other (often rare) ephemerals of disturbed ground on heathlands include three-lobed water crowfoot Ranunculus tripartitus, slender centuary Cicendia filiformis, chaffweed Anagallis minima, and allseed Radiola linoides. All require trampling in the winter to create bare/poached ground but may suffer from trampling during spring and summer. Dwarf rush Juncus capitatus a winter annual germinating in the autumn (present only on Anglesey in Wales) may suffer from trampling in winter, although it too requires bare open condition to germinate (possibily maintained by standing water in winter and/or drought in summer).

Bare ground or sparse vegetation cover over suitable sandy soil are highlighted by Alexander et al. (2005e) as the habitat for a large number of bees and wasps, including some species that occur in Wales, such as the nomad bee Nomada signata and the solitary wasp Methocha articulata. Bare ground provides a range of functions, including nesting sites, foraging habitat and also basking surfaces for thermophilic species (Stubbs and Drake, 2001).

Heavy use of sandy tracks, particularly by horses or mountain bikes, causes the sand to be loose and continually disturbed, rendering the habitat of low value to many invertebrates (Symes and Day, 2003). Species that burrow into flat surfaces such as the centres of paths are likely to be particularly vulnerable, as loose sand may not support their burrows and the churning may make it impossible for them to relocate their burrows once dug. The friable nature of heathland soils makes them particularly vulnerable to these trampling impacts whilst an additional potential impact from horse riding is nutrient enrichment of heavily used tracks, which may lead to a change in plant species composition (Liddle and Chitty, 1981, Liley and Clarke, 2002, Liley et al., 2002) – see the contamination section.

Horse riding has been implicated as a major factor in the severe decline of the heath tiger beetle Cicindela sylvatica in England (the species is absent from Wales). Horses‘ hooves churn up sandy heathland tracks making the habitat unsuitable for the beetle. Excessive disturbance from horse riding and motorcycle activity has also been identified as a threat to

1 http://www.plantlife.org.uk/uk/assets/saving-species/saving-species- dossier/Lycopodiella_inundata_dossier.pdf

34 a number of aculeate Hymenoptera that occur in Wales, including the UKBAP Priority Species such as the cuckoo bee Nomada ferruginata, shrill carder bee Bombus sylvarum, and a ruby tailed wasp Chrysis fulgida, (UK Biodiversity Group, 1999b, UK Biodiversity Group, 1999a, Alexander et al., 2005e).

Heavily trampled ground can also present a barrier to some invertebrates (and possibly even small mammals in some habitats). Carabid beetles have been shown not to cross tracks or habitat different from their chosen type ( Mader, Schell and Kornacker (1990) and Lindsay (2001) cited in Penny Anderson Associates (2006)). Similarly a species of snail was shown not to cross a 3m wide track dividing its habitat, even though travelling that distance was quite possible (Baur and Baur, 1990). This isolation of habitat fragments could have implications for small populations, for instance in maintaining genetic diversity. Where access results in the increase in path widths or new desire lines further fragmentation can occur within sites.

Lower plants – lichens and bryophytes – can be a very important component of heathland and these too can be adversely affected by trampling. Most of the studies relate to upland moor and mire communities or dune systems but the studies suggest that dry heathland bryophytes and extensive carpets of terricolous lichen communities, especially those with reindeer lichens (Cladonia arbuscula, C. portentosa etc.) which occur on certain heaths, and occasionally on dunes, could be irrevocably damaged by excessive trampling. This effect in dry conditions, when the lichens are at their most brittle, is most damaging (Liley et al., 2002, Liley and Clarke, 2002). Certain rare species, such as cliliate strap-lichen Heterodermia leucomela,2 and golden hair lichen Teloschistes flavicans 3 species recorded from maritime cliffs, slopes and coastal heathland sites, are thought to be potentially vulnerable to trampling e.g. when occurring adjacent to heavily used paths

Upland and Montane Habitats As with the evidence for trampling damage to heather-dominated vegetation on lowland heaths, upland heather moorland is also vulnerable (Bayfield, 1979). Indeed the damage may be more severe as recovery is slower under cooler conditions, with the possibility of further damage to bruised stems from frost (Watson et al., 1966). Increasing altitude tends to exacerbate this trend with recovery rates even slower in alpine or sub-alpine conditions (Whinam and Chilcott, 2002).

The levels of trampling responsible for damage routinely reported in the literature are quite low and could be expected on many mountain/moorland areas in a single season. The vegetation growing over shallow peat soils seems in general more vulnerable than that on mineral soils, with grass-dominated cover on mineral soil more resilient than old heather

2 Anecdotal evidence on Plantlife website: http://www.plantlife.org.uk/uk/plantlife-saving-species- under-our-care-lich-heterodermia-leucomela.htm

3 http://www.arkive.org/golden-hair-lichen/teloschistes-flavicans/info.html

Calluna vulgaris, bilberry and crowberry on peat soils. Deer grass Trichophorum cespitosum is characteristic of trampled swards. Surface consolidation (decreasing aeration) along paths may favour this species (Shaw et al., 1996, Rodwell, 1991).

Studlar (1980) noted that bryophyte cover, frequency and species richness was greater on trails in the Appalachians than in the adjacent vegetation, possibly due to the reduction in litter and competition with taller vascular plants. The importance of the impact of trampling will, therefore, relate to which species are involved, and whether the key species are those that are more tolerant of trampling and can take advantage of the more open conditions. Other studies have highlighted the more vulnerable species. Racomitrium lanuginosum, (an important moss component of high level montane heath in UK) for example, was found to be very sensitive to trampling in Iceland (Jónsdóttir, 1991).

Bryophytes, therefore, vary in their tolerance to trampling. Studlar (1980) suggests that those with thick midribs, short, tough concave leaves, and protected meristems are more likely to show resilience to trampling. Some pleurocarpous mosses such as Hypnum spp (with a weft-like growth form rather than upright shoots) are rather more resilient. However, Hamberg et al., (2008) found (in forests in Finland) particularly vulnerable species included Brachythecium spp., Pleurozium schreberi, Hylocomium splendens, (all three typical moorland pleurocarps) and Dicranum polysetum, while bryophytes that grow in low, thick tufts, such as Pohlia nutans, may be more resistant to trampling than the longer, looser shoots of Brachythecium species (Hamberg et al., 2008).

A number of studies report negative effects of rock climbing on montane cliff vegetation (e.g. Nuzzo, 1996, McMillan and Larson, 2002, Rusterholz et al., 2004). McMillan and Larson (2002) showed that the proportion of introduced plants was significantly greater on climbed faces. Bryophyte diversity and richness was significantly lowered on climbed rock while lichen frequency was the same, though the species richness was much less and community composition was affected. Muller et al (2004) found in Switzerland that community composition was affected on climbed slopes and specialised rock species of plants occurred less frequently. However a Canadian study (Kuntz and Larson, 2006) suggested that some of the comparative difference between climbed and pristine rock faces was related to choice of climb (with some selection of climbs favouring cleaner, less vegetated routes) and microclimatic differences.

Bog Lowland mires and upland blanket bog are important habitats in Wales and both habitats depend on the peat-building and water-holding properties of the principal components, bog mosses Sphagnum spp. These are especially vulnerable to trampling damage and various studies have shown that most types of Sphagnum dominated mires could be degraded by persistent trampling and erosion. Bayfield (1971) found only 80 tramples across Sphagnum in Cairngorm destroyed the plants (with no signs of recovery after 23 months). In the different climate of the Appalachian mountains in Virginia, Studlar (1980) found that 130 passages reduced Sphagnum recurvum to a few isolated leaves and stem and root

36 fragments. Bryophytes create the micro-habitat suitable for other species such as grasses and sedges, therefore their loss means tha the conditions for growth are changed. Other sensitive species include bog myrtle Myrica gale and butterwort Pinguicula sp.4

Low levels of trampling may increase species diversity, with species such as sundew Drosera spp. colonising. Studying trampled blanket bog in Ireland, McGowan (2002) found that some of the invading species of the sites with minimum levels of trampling were typical of heathland, and acid grassland habitats, or nutrient enriched sub-habitats on intact blanket bogs. Up to 20 species can invade trampled bogs, including rushes Juncus and Luzula spp. and bent grasses Agrostis spp. which can colonise bare, eroded peat although unable to survive the severest levels of trampling. Bell heather Erica cinerea, a plant is more commonly found on the drier, heathy habitats is an indicator of a change to drier conditions on the damaged bog surface.

On the more damaged areas of medium trampling, McGowan found that the diversity of species had decreased. There was considerable bare ground on paths, whereaspath edges were often marked by black bog rush Schoenus nigricans, bog myrtle and deer grass Trichophorum caespitosum. Mosses, liverworts and lichens were recorded with both low frequency and cover, and were generally confined to the path edges where suitable microhabitat was available at grass and sedge tussock bases. These trampled paths act as rudimentary drainage channels, with water running over the eroded areas of bare peat.

In the areas of most severe trampling damage the vegetation cover was sparse. Where there was severe trampling damage, vegetation structure was reduced and there were few if any flowering stems produced. Tufted plants, such as deer grass and black bog rush, were better able to survive in severely damaged areas, than were the heathers, whose erect branches and apical growth tips were too sensitive to the trampling damage.

Some rare vascular plants can also occur within Sphagnum carpets, such as the globally rare bog orchid Hammarbya paludosa and like Sphagnum, this is vulnerable to trampling (Lake et al., 2001).

Fen and swamp There is comparatively little research material on the effects of trampling on fen vegetation, perhaps because the habitat is to some extent self-protecting due to its general wetness and usually small, discrete location. Unlike the far more extensive mire communities of upland moors for instance, fen habitat can be avoided rather than traversed. The taller components in wetter conditions, such as reed Phragmites communis and reed canary-grass Phalaris arundinacea are likely to be more vulnerable to trampling than shorter turf on drier soils (Rees and Tivy, 1977). A five-year controlled trampling study (100 passages/year, all in August) in four rich fens in central Norway showed severe damage (Arnesen, 1999) . Vegetation cover was reduced, woody species and herbs disappeared or were reduced in

4 http://www.ipcc.ie/infotrampling.html

37 cover. However some sedges, marsh horsetail Equisetum palustre, common cottongrass Eriophorum angustifolium and certain bryophytes appeared to be quite tolerant. Sphagnum was vulnerable however. Recovery was monitored over 13 years and, even after this time, there were fewer species and less vegetation on the tracks. The trampling also resulted in up to c. 7cms reduction in the height of the ground surface, which consequently flooded more regularly.

Some degree of light trampling may not be deleterious, and may open the closed tall cover and providing niches for less competitive plants, or edge habitat for invertebrates and birds, but the Norwegian study shows how even light effects can cause quite substantial change and for a long period. Stammel and Kiehl (2004) explored the importance of hoof prints for species recruitment in fens. Their results suggest that on wet peat soil the negative effects of soil compaction and changed water and light availability probably prevail over positive effects such as the promotion of subordinate species.

Standing water/rivers Many of the component species and the general characteristics of the plants of waterside habitats are in common with those of fens, but the feature of open water – lakes or rivers – makes this a focal point and preferentially targeted. Trampling effects can therefore be localised and especially intense along the margins of the water feature.

Under heavy recreational pressure 31% of the reed swamps along the Havel River in West Berlin disappeared over a five year period with riverside access, (Sukopp, 1971). At first, the disturbance allowed space for ruderal species, especially where meadows lay adjacent to the river, but the reed stands vanished with intensive use and the river margins then eroded. On paths beside Scottish loch shores (Rees, 1978) common reed, reed canary-grass and reed sweet-grass Glyceria maxima were replaced by bent grasses Agrostis spp, meadow-grass species Poa spp., amphibious bistort Periscaria amphibia and knotgrass Polygonum sp., i.e. species found elsewhere to be general tolerators of trampling. Heavily used paths were reduced to bare mud.

Boat traffic can result in damage to aquatic plants. For example, the principle threat5 given for the BAP priority floating water plantain Luronium natans is the re-opening of waterways, with subsequent high levels of motorised recreational boat traffic. This can directly suppress growth of the plant through increased turbidity of the water.

An indirect effect, from heavy recreational pressure on adjoining land, was noted in the mountains of central Spain, where environmental conditions started to decline in the 1970s as a result of increased visitor numbers, initiating severe soil erosion processes and increasing nutrient load in Peñalara Lake. This combined effect altered the composition of aquatic biological communities (Toro and Granados, 2002).

5 http://www.ukbap.org.uk/UKPlans.aspx?ID=427

The risk of transfer of alien invasive species, on boots for instance, is noted though as yet appears un-researched. The rapid spread of New Zealand pygmy-weed Crassula helmsii is well known and may be due in part to the incidental effects of trampling in wetland habitats, as well as transfer by wildlife.

Sand dunes The trends in vegetation response to trampling on sand dune habitat are broadly similar to those found in grassland; the more stressed the environment and unstable the substrate, the greater the impact. Thus, fore dunes with marram Ammophila arenaria may be very susceptible to trampling, while rank grasses and dune heath are moderately susceptible and short turf and scrub most resilient (Boorman and Fuller, 1977). In unmanaged dune grassland, trampling results in a progressive decline in height of vegetation and less litter; and also some increase in pH associated with compaction (Slatter, 1978). As with other grasslands, the increase in available phosphorus noted (Milwain, 1984) could be directly linked to the use of paths for dog walking.

Coombes (2007) explores the relationship between the amount of passes (footfalls) and reduction in vegetation cover in different soft coastal habitats. For most habitats (yellow dunes, grey dunes and saltmarsh) the relationship appears to be linear, suggesting that the impact is proportional to the amount of access. The slope is steepest for yellow dunes and shallowest for saltmarshes, suggesting that of these, yellow dunes are the more sensitive. The relationship for foredunes appears—uniquely among the habitats assessed—to be curvilinear, with a small amount of trampling resulting in a disproportionately high impact.

Again, as for other grasslands, some light trampling in otherwise unmanaged dune grassland may benefit less competitive plants such as some annuals and invertebrates, but the habitat is very prone to erosion and the creation of increasingly wide, bare pathways. It is generally accepted that recreational pressure results in a decrease in species diversity within dunes (Bonte and Hoffman, 2005), and that a threshold can be reached where irreversible damage can occur (Curr et al., 2000, Ritchie, 2001, Covey and Laffoley, 2002), although it is often difficult to identify at what point this may threshold may occur.

Moderate trampling in the lichen-rich grey dune vegetation of a Danish dune system (Christensen and Johnsen, 2001) resulted in a reduction in species number and vegetation cover; a community of colonizers or early succession species now predominates. In the most trampled parts, only a few (or even no) species remain, and the substrate is worn down to bare sand. Frequent trampling means that later successional stages become an increasingly rare element of the vegetation. Moreover, the invasion of exotics (including heath star-moss Campylopus introflexus) and atmospheric nutrient enrichment also threaten native lichens at the Danish site.

Some species may benefit, at least from light trampling that maintains more open vegetation or some bare ground. The belted beauty moth, an endemic subspecies found only in Britain, and confined now to just three coastal sites, one on dunes in Wales, requires light trampling to maintain more open, early-successional herb-rich dune grassland and

39 prevent it from becoming rank (Howe et al., 2004). Similarly, a number of ground beetles (Carabids) may benefit from controlled public access on sand dune systems. The tiger beetles Cicindela hybrida and Cicindela maritima are often associated with sandy pathways (Luff, 1998, Telfer and Eversham, 2000): it is possible that the correct level of trampling can create suitable conditions (ground which is neither too heavily compacted nor too loose and churned up) for larval burrows (Alexander et al., 2005b).

It is clearly difficult to generalise between species, groups and sand dune sites. Moreover, for many sand dune invertebrate species it may be difficult to untangle the importance of access levels in modifying habitats from the direct impacts of trampling and disturbance. Work in Belgium on the digger wasp Bembix rostrata (Bonte, 2005) found that trampling from people and cattle impacted both directly by interfering with larval feeding and destroying nest sites, and indirectly by causing the deterioration of nest site quality by changes in the structure of the soil surface. However, Bonte suggests that it is not possible to make generalisations between species.

Shingle Whereas for sand dune vegetation some degree of light trampling can be beneficial for some plants and invertebrates, this appears virtually never to be the case for shingle habitats. The shingle survey of Great Britain covers a number of sites in England and Wales (Sneddon and Randall, 1989). At many of these, trampling was noted as one activity causing damage to fragile shingle vegetation (Sneddon and Randall, 1993a, Sneddon and Randall, 1993b, Sneddon and Randall, 1994). Damage to ridge structures is also an issue. One of the main causes of damage is the breaking up of the surface layers of vegetation and the fine humic layer that may take many years to be deposited. As a result, damage to vegetation may not be possible to reverse. Spokes (1997) studied shingle vegetation and trampling and compared data from 1991 with that collected in 1997 on a shingle habitat at Slapton in Devon. The results indicated that untrampled areas were more diverse than the trampled areas. Hewitt (1973) came to the same conclusion on Chesil Beach in Dorset. Communities with abundant lichens are susceptible to trampling, again particularly during dry weather. A single pass may be sufficient to cause irreparable damage (Doody and Randall, 2003) .

Disturbance of habitat from trampling or vehicles also has a negative impact on the majority of shingle invertebrates (Shardlow, 2001), an assessment supported by Kirby (2001): ―”public access to shingle habitat is probably always damaging to some extent. Shingle communities are easily damaged by trampling because of the instability of that habitat”. Off- road vehicle access has been identified as a threat to invertebrates on some sites (Alexander et al., 2005c).

Rocky shores A number of studies, from various locations across the world have investigated the effects of trampling on rocky shores. These generally have shown that trampling causes a reduction in

40 cover of a range of algae species. The extent of damage increases with the intensity of trampling. Increased intensity of trampling leads to increases in the amount of ephemeral algal species and extends recovery times from months to years (Fletcher and Frid, 1996, Keough and Quinn, 1998, Schiel and Taylor, 1999, Beauchamp and Gowing, 2003, Milazzo et al., 2004, Irvine, 2005). The extent of damage and removal does not seem to be affected by the hydration state of the algal mat. Where the plant cover is not completely removed, effects of trampling can cause loss of vesicles or air bladders and reproductive structures (Keough and Quinn, 1991, Denis, 2003).

Trampling can also cause changes in species diversity (Pinn and Rodgers, 2005, Van de Werfhorst and Pearse, 2007) with a decrease in larger branching algae and an increase in smaller emphemeral and crustose species (Pinn and Rodgers, 2005), at least during the summer (Fletcher and Frid, 1996). Similarly desiccation following the removal of the brown alga Fucus canopy can lead to a reduction in coralline algae (Keough and Quinn, 1998, Schiel and Taylor, 1999). Jenkins (2002) found no reduction in turf-like algae, but experimental trampling levels within the observed range of trampling by visitors caused losses of turf-like algae in another study (Brown and Taylor, 1999).

Reduction in algal cover and increase in bare rock can continue after trampling has stopped (Jenkins et al., 2002); effects can be variable depending on the degree of summer desiccation (Keough and Quinn, 1998); and communities damaged in autumn can take longer to recover than those damaged in spring as recovery is greatest during the spring and summer (Schiel and Taylor, 1999).

There seem to have been few studies on the effects of harvesting on rocky shores. The main species are shellfish, crustacean (such as peeler crabs) and sea urchins for bait or food, and the general effect of over-harvesting is to reduce populations and remove old adults; there can be consequences for reproductive success (Murray et al., 1999) and changes in species dominance/community composition (Duran and Castilla, 1989). Harvesting (as well as other activities including educational trips) can involve turning over rocks. This can be particularly damaging as organisms can be crushed when the rock is turned over or put back, constant turning of the same rocks can result in so called ‘monk’s head rocks’ with only a fringe of organisms around the edge (Addessi, 1994). There is a long tradition in the UK of rock pooling by children, which serves as a way of educating them about shore wildlife. There is no evidence that the rocky shore flora and fauna around our coasts is being significantly damaged by this recreational activity. Research does suggest that visitors to rocky shores who had a greater knowledge of intertidal ecology were more likely to have a greater impact than visitors with less knowledge (Alessa et al., 2003).

Maritime cliffs Hewitt (1973) description of the vulnerability of cliffs includes climbers removing vegetation; the erosion of soft rocks where visitors scramble down to the beach causing gullies can develop, and the loss of vegetation and erosion on cliff tops, especially at view points. He gave examples at Penrhyn Wawr, Holyhead; Elegwg Stacks, Pembrokshire; Portland Bill,

Dorset; and Beachy Head, Sussex where vegetation had been lost, soils eroded and the exposed rocks became polished and worn. Seepages, streams and gullies are often important sites for invertebrates and plants (Telfer, 2006, Whitehouse, 2007) and these may be targeted as routes down the cliff, causing erosion and thus loss of interest.

Cliff climbing, like inland rock and mountain climbing, can result in damage to vegetation, though choice of site may initially select less vegetated routes. This may not apply to lichens however, which could go unnoticed in a preliminary assessment of climbing suitability. In cliff situations in Canada, Farris (1998) noted that larger foliose and fruticose species of lichen were especially vulnerable to crushing and trampling.

Saltmarsh Saltmarshes and particularly mudflats do not lend themselves to easy access and therefore have a degree of self-protection. Comparative studies of trampling impacts on different coastal habitats indicate that saltmarsh is the most resilient habitat relative to sand dunes, coastal grasslands etc. (Andersen, 1995, Lawesson, 1998, Coombes, 2007). However, where trampling does occur it may still have significant effects. At Lindisfarne, a study of the impacts of trampling on the tidal flat infauna on a regularly used Pilgrim route (Chandrasekara and Frid, 1996) concluded that trampling caused changes to the infaunal community. He also indicates that the vegetated saltmarsh has developed a permanently distinguishable path along the route and the vegetation composition may have been altered.

Another recent study has linked the decrease in the area of saltmarsh in parts of Australia to access (Laegdsgaard, 2006). Even annual visits to fixed sample points can cause visible changes to the vegetation (Boorman, 2003). The marshes which are the most liable to damage from trampling are those with poorly drained soils made up of fine sediments (Boorman, 2003). Pioneer saltmarsh may also be particularly vulnerable; this habitat, at Holkham, is crossed by people visiting the beach, who fan out after crossing the dunes near a main access point. A clear network of bare paths is visible (Liley, 2008).

A study (Turner, 1987), conducted in cord grass Spartina-dominated saltmarsh in the US, sought to simulate and compare grazing and trampling by cattle. The results suggested that trampling may have a greater impact - trampling by horse riders may have a similar effect.

Trampling damage can directly affect saltmarsh vegetation. Any damage to the vegetation cover of the saltmarsh carries with it the risks of erosion damage over a much wider scale. It will also have an impact on the soil fauna with possible consequences for the functioning of the marsh ecosystem as a whole (Boorman, 2003).

Lagoon snail Paludinella littorina may also occur in saltmarsh and is sensitive to physical disturbance of the habitat caused by access (Killeen and Light, 2002). Some mild trampling however, exposing areas of bare mud may be of benefit to some flies associated with saltmarsh; these include the cranefly Dicranomyia complicata (RDB2) (Alexander et al., 2005a).

Saline lagoons Trampling in the usually shallow water of saline lagoons can be an incidental effect from other activities such as bait digging, shellfish collection and recreational activity such as sailing and windsurfing. The impacts on marginal vegetation are likely to be similar for shingle and saltmarsh; while impacts on invertebrates will be comparable to those on mudflats. Activities can have also have indirect effects such as soil erosion from people, horses and vehicles, which causes run-off with subsequent turbidity in adjoining lagoon water columns.

Mud & sand flats Trampling of mudflats leads to changes in invertebrate community structure (Chandrasekara and Frid, 1996) both directly (addressed in Section 67 disturbance) and indirectly through habitat change. Trampling alters the surface topography of the mudflats, which can indirectly affect recruitment and spatial distribution of microalgae (Wynberg and Branch, 1994) and macrofauna (see Rossi et al. (2007) and references therein). Furthermore, compaction of the sediment might alter the exchange of nutrients and oxygen between the sediment and the overlying water, change sediment accumulation rate and, again, modify population dynamics and distribution of animals in the mudflat (Rossi et al., 2007).

The reduction in invertebrates on tourist beaches is likely to be a result of direct mortality from trampling and also through habitat modification, which on many heavily used beaches may include beach cleaning. A study of coastal systems in the Mediterranean and Baltic (Gheskiere et al., 2005) shows that upper beaches heavily used by tourists are characterised by a lower proportion of total organic matter, lower densities of invertebrates, lower diversities (including absence of certain insects and nematodes) and higher community stress compared to nearby non-tourist locations. The proportion of total organic matter was found to be the single most important factor for the observed differences in the faunal assemblage structure at tourist versus non-tourist beaches.

In both sand- and mud-dominated habitats, the intensity of the trampling is important (Chandrasekara and Frid, 1996, Moffett et al., 1998), with relatively little damage at low trampling intensities.

Bait digging can negatively impact upon benthic invertebrate communities both directly through the collection of bait and also through the physical damage to the habitat. Relatively immobile species, such as the cockle Cerastoderma edule, can be easily dislodged and buried by digging and may not survive due to asphyxia. Declines in the cockle population in the 1950s and 60s on the North Norfolk coast were found to be due to digging for bait worms (Jackson and James, 1979). A study by Farrell (1998), described a reduction in numbers of cockles in experimentally dug areas of Chichester Harbour. In the same study the large sedentary worm, Amphitrite johnstoni and Harmothoe imbricata (which share the same burrow) were completely lost from the experimental plots.

We summarise impacts by habitat, according to season, in Table 5.

Table 5: Summary of seasonal vulnerability to damage, by habitat Spring Summer Autumn Winter

 Shade adapted broad-leaved herbs and bulbs vulnerable  Can cause soil compaction – affects soil fungi and thus tree 1 Broadleaved, mixed and yew  Initial trampling most damaging. health woodland  Trampling has greater effect than flower picking  Can cause soil compaction – affects soil fungi and thus tree health  On PAWS6 broad-leaved herbs more vulnerable than grass or  Bryophyte & lichens vulnerable 2 Coniferous woodland shrubs  Bryophyte & lichens vulnerable 3 Boundary and linear features No literature found 4 Arable and horticulture No literature found 5 Improved grassland  Effect on invertebrates of litter integration (no discrimination made between seasons). Can occur before vegetation shows impacts

 Unmanaged grass (not cut/ grazed) shows most effect  Unmanaged grass (not cut/ grazed) shows most effect  Moderate levels of trampling may result in higher spp 6 Neutral grassland diversity  Moderate levels of trampling give higher spp diversity  Resilience likely to be greatest in growing season  Bare ground inverts can benefit from light trampling  Bare ground inverts can benefit from light trampling 7 Calcareous grassland See above

8 Acid grassland See above

6 PAWS – plantations on ancient woodland sites

9 Bracken  Early stages of trampling likely to have biggest effect,  Trampling may increase frost damage to rhizomes trampling more damaging when plant above ground  Heather vulnerable.  Damage worse in cool/frosty weather  Dry heath most resistant  Some spp of invertebrates and non-competitive plants 10 Dwarf shrub heath  Some spp of invertebrates and non-competitive plants benefit from light trampling benefit from light trampling  Dry lichens more vulnerable than wet  Deer grass characteristic of trampled swards  Tall sward more vulnerable 11 Fen, marsh and swamp  Decrease in vegetation cover, woody species and herbs  Some sedges, horsetail and cottongrass may be more tolerant  Sphagnums most vulnerable bryophytes  Sphagnums always very vulnerable, heather, cottongrass and purple moor-grass more resilient  Erosion likely to be 12 Bogs  Light trampling may increase species diversity (e.g. Sundews). greatest in winter  Rushes, bent grasses and bell heather may increase under moderate tramping 13 Standing open water and  Tall fen marginals most vulnerable canals  Light trampling may increase species diversity, heavy trampling will result in loss of fringing reeds

14 Rivers and streams As above

15 Montane habitats  Vegetation on peaty soils more vulnerable than on mineral soil  Frost effect can add to  Some bryophytes very vulnerable mechanical damage  No literature documents 16 Inland rock  Diversity of bryophytes and vascular plants less on climbed routes (may be due to climber choice seasonal effects, but life- of route) cycles of plants suggest impact less in winter 17 Built-up areas and gardens No literature found

1 Saline lagoons  Impacts on marginal vegetation similar to salt marsh/shingle

 No literature documents 2 Coastal vegetated shingle seasonal effects, but life-  Effects of trampling adverse and long-lasting cycle of plants suggest impact less in winter 3 Rocky shores  Trampling leads to loss of algal cover  Trampling leads to loss of algal cover  Faster recovery in growing season  Gullies/flushes may be very vulnerable to erosion – no 4 Maritime cliffs and slopes  Gullies/flushes may be very vulnerable to erosion seasonal differentiation in literature but erosion likely to be  Larger lichens very vulnerable worse in autumn/ winter  Larger lichens very vulnerable 5 Saltmarsh  Less vulnerable than dune or shingle but can be affected.  Less vulnerable than dune or shingle but can be affected Faster recovery likely in growing season 6 Mudflats and sandy beaches  Compaction of substrate – may affect fauna  Changes to microtopography affect micro-algae and fauna  All dunes affected , fore  All dunes affected , fore dunes most vulnerable.  All dunes affected, fore  All dunes affected, fore dunes most vulnerable  Faster recovery likely in dunes most vulnerable dunes most vulnerable  Faster recovery likely in growing season  Some light trampling of tall  Some light trampling of tall growing season  Some light trampling of tall grass fixed dune may grass fixed dune may 7 Coastal sand dunes  Some light trampling of tall grass fixed dune may increase species diversity increase species diversity grass fixed dune may increase species diversity of both plants and inverts of both plants and inverts increase species diversity of plants and inverts

of plants and inverts  Seasonal impact can be very adverse for inverts – damage to nest holes

Fire

Introduction A distinction needs to be made between a controlled fire and a wild fire. A controlled burn is a deliberately started fire, with the intention of burning off excess or old vegetation in a controlled or managed way in a defined area (Gimingham, 1992, Webb, 1997). A wild fire is a fire which is not controlled and can spread in an unpredictable way until it goes out or is put out. Such wild fires may be caused by a controlled fire getting out of control, or they may be started accidentally or deliberately by arson. Whereas controlled fires (and those accidentally derived from them), are started during the winter months (in the legal burning season), wildfires are more likely to occur in dry periods in spring, summer and autumn.

Wildfires can be caused accidentally from discarded cigarettes, by sparks from a campfire, BBQ or from burning a dumped or stolen car, from fireworks, as a result of a controlled fire getting out of control, from discarded bottles in strong sunlight, from children playing with matches or similar, and from deliberate arson (Anderson, 1997);Tucker 2003).

In Wales, in the last year for which full figures are available (2006/7) there were 7238 grass (includes heath and moor) and forest fires in Wales with 63% of these in South Wales, many close to the built up areas in Rhondda Cynon Taff and Caerphilly. The South Wales Fire and Rescue Service believe some 95% of these fires were started deliberately7. The mid and West Wales and South Wales Fires and Rescue Services, together with the Forestry Commission have begun a research project over two years to explore the motivations of those who start fires and to assess methods used to prevent such behaviour. The project started in January 2009 and a scoping exercise has been completed 8.

Based on 217 questionnaires from a sample of lowland heaths in Dorset, Kirby and Tantram (Kirby and Tantram, 1999b) found that 61% of fires were caused by arson, 8% from management fires getting out of control, 7% from bonfires and the remainder from camp fires, burning refuse, vehicle fires, property fire and sparks from a railway. The only natural cause of fire was from lightning. The same study noted that there was a widespread belief among the public and nature conservation professionals that most fires were deliberate and that children were often believed to be responsible (this would be most relevant on sites close to residential areas rather than remote uplands).

A number of studies have linked the incidence of fires with areas used by the public, or with the extent of urbanisation. In the Peak District National Park during 1970-1995, 84% of 324 recorded fires were next to roads, paths or within areas of open access, and many burnt areas on Exmoor are close to public roads (Miller and Miles, 1984). Kirby and Tantram (1999a) noted that of the 26 lowland heathland SSSIs in Dorset with the highest number of fires, 1990-1998, 70% were located in or adjacent to urban areas, including the top nine. The highest number of recorded fires was at

7 Women.timesonline.co.uk

8 [email protected]

Canford Heath SSSI (adjacent to Poole, Dorset), with a maximum of 12 fires in a day and 97 in a month. They also found that more fires occurred on heathland SSSIs with more densely developed urban areas within 500m of their boundaries, and considered that a possible threshold was operating such that fires were more likely to occur on sites with more than 15% of the land developed within 500m of their boundary than on sites where adjoining development was below this level. A study or recreational fire rings used for cooking found that these tended to be near streams, away from forest roads and close to open spaces in the forest (Hegetschweiler, 2008)

Kirby and Tantram (Kirby and Tantram, 1999b) found that during the period1990-1998, wildfires could occur at any time of year but were most common between April and August. Fires were more likely to occur at weekends than weekdays (a finding also noted by Anderson (1986) in the Peak District), during school holidays than term time, and during the afternoon and early evening than at other times of day. A possible explanation for this latter finding is that this is the period when children come out of school and working parents have not yet come home. In another study on urban heaths Murison (2007) found that 77% of wildfires occurred between April and September, with the peak wildfire frequency in April with 23% of the fires. Fires on the higher moorlands away from urban areas do not share this daily occurrence pattern due to the greater effort and time needed to access them, and the predominance of daily visits by recreational visitors rather than school children.

A further analysis of these data, and a further data set for lowland heathland for 2003 by Rose and Clarke (2005) showed that the number of wildfire incidents varied widely between years. Fires were more common in the summer months than winter months and there was no evidence that fires were more or less common in 2002-3 than in 1993-8. They also found that generally there were fewer fires between Tuesday and Thursday than between Friday and Monday and that the median start time for fires was between 16.00 and 17.00 hours whether or not it was term time.

Fires can result in medium or long-term damage to ecosystems. The effect of a fire depends on its intensity, which is determined by the heat generated and the period of time the heat is applied. A hot fire moving slowly over the ground can cause far more damage than a superficial fire moving quickly. Factors which influence the intensity of a fire are the pattern, quantity and combustibility of the vegetation, the prevailing weather conditions particularly relative humidity, ambient temperature and wind and the topography of the site (Kirby and Tantram, 1999a, Tantram et al., 1999). Some of these factors particularly vegetation patterns, wind direction and speed and topography can also determine the extent of a wildfire in the event that it is not put out by human intervention.

An analysis of weather conditions by Kirby and Tantram (1999b) found that temperatures, sunshine, relative humidity and wind direction were important in predicting the likelihood of wildfires. They suggested that the risk was high when, in a given 10/11 day period, the number of sunshine hours reaches 100, cumulative relative humidity is<600% and there has been <5mm of rainfall overall. Anderson (1986) recorded that the number of fires in the Peak District National Park were nearly 80 times higher in the very dry year of 1976, (following a drought which had begun in May 1975), than in the wetter year of 1973. She also found that on the moorlands, compared with lowland heathland, the incidence of fires increased substantially after a month of below average rainfall, and with concomitant high temperatures.

In the UK, most studies on fire have been carried out on lowland heath or moorland. This is because fire has been a standard management technique on these habitats for centuries and is still practised for commercial reasons, and much of the vegetation is therefore fire adapted. Many of the studies concern managed winter fires. In practice, as vegetation is usually drier in summer than winter, summer wildfires may consume more of the vegetation than would normally be burnt in a controlled fire. Summer wildfires can be hotter and more intensive than controlled winter fires, removing litter, destroying vegetation and burning down into organic soils.

In Dorset, Webb (1997) estimated that in 1959, 8.1% of the heaths were burnt, while a series of surveys in 1978, 1988 and 1998 found that 12% (945ha), 4.8% (382ha) and 1.2% (85ha) respectively of the heaths had been burnt, strongly suggesting a reduction in total areas burnt over the previous 30 years (Webb and Haskins, 1980, Webb, 1990, Rose et al., 2000);. A reduction in the number of fires (58) was also recorded on Yateley Common to Castle Bottom Common in Hampshire by (Liley, 2004), compared to an earlier survey (76 fires) by Hall (Hall, 1996). The median size of fire at Yateley in 2004 was 0.0009ha with only five fires above 0.1ha, and two above 1ha and no obvious proximity of fires to car parks. In studies in Dorset, (Rose and Clarke, 2005) reported that, of 203 fires in 2002/3, only six burnt more than 1ha, 83% of fires were smaller than 0.1ha, and one fire, of 60ha represented 62.1% of the total burnt area, and Murison (Murison, 2007) found that average wildfire size was 0.81 ha with 96% of burns covering less than 0.5 ha and only two of 103 fires exceeding 10ha in size..

In the Peak District National Park Anderson (1997) recorded the numbers of annual fires 1970-1995 between 1 and 79, ranging from less than 0.1 ha to 1210 ha in size. There was no apparent reduction in average size or numbers of fires over the period despite considerable efforts to control this. McMorrow in Glaves et al (2005), noted that there were more uncontrolled fires in the Peak District in April (35) than September (9) during 1971-2000. A similar pattern has been reported for Dartmoor (Kirkham in Glaves et al 2005). February to June are the drier months in the Peak District, and combined with higher temperatures later in spring would account for this pattern.

The heat and variability of the temperature in a fire is partly a function of the age and structure of the vegetation which can determine the fuel load (e.g. dry grasses, including reed, other herbaceous plants, scrub, heather Calluna vulgaris and other woody plants). Following a fire, regeneration will depend on the character of the pre-burnt vegetation.

As the structure and composition of the vegetation affects its flammability, previous management that removes vegetation such as cutting or grazing, will affect fire temperature, intensity and spread.

A normal, controlled fire in upland heath will reach an average temperature of between 400-600°C, which will be sustained for about two minutes as the fire passes through the vegetation (Mallik and Gimingham, 1985) In grassland, average temperatures were higher, particularly in tussocks at between 560-805°C, and lasted about one minute except where tussocks were dense (Lloyd, 1968). Recorded temperatures for different habitats are shown in Table 6. Apart from the lowland dry heath measured by (Allchin, 1997), where fire temperatures were lower, in most habitats a maximum fire temperature of about 8-900°C was attained regardless of habitat. Temperatures were particularly variable in the blanket bog. Fires can also back burn and change direction with wind eddies, or small scale changes in the flammability of the vegetation, thus providing significant variation within a single event. Fires can, therefore, re-burn areas, and penetrate down into the

50 peat, possibly burning here for days or weeks. Where there is a considerable quantity of inflammable material fires can be very intense with flames reaching 10m in height in gorse fires, and 5m in reed fires (George, 1992).

Temperatures at the ground surface can be as high or higher, than in the canopy of heather, but is lower at the soil surface in grass fires (85-145°C was recorded by Lloyd 1968 in fires on limestone grassland). Normally, 1cm below the soil surface, the temperature will remain below 100°C (Hobbs and Gimingham, 1984b). The effectiveness of the litter layer as an insulator will depend on its thickness and moisture content (Valette et al., 1994). Fires in standing reed can leave an unburned stubble and an only slightly scorched litter layer (Hawke and Jose, 1996).

Table 6: Fire temperatures in vegetation canopy from different habitats Habitat Mean °C Min – Max °C Reference

Dry heath, N-E Scotland <400 220-840 (Whittaker, 1961)

Dry heath 670 -940 Kenworthy in (Webb, 1986)

Dry heath, N-E Scotland 340-790 (Hobbs and Gimingham, 1984b)

Dry Heath, Dorset, Devon 316 173-581 (Allchin, 1997)

Blanket bog N-W Scotland 103-886 (Hamilton, 2000)

Wet heath W Scotland 670-800 (Currall, 1981)

Limestone grassland 560-805 (Lloyd, 1968)

On dwarf shrub heath, heather regenerates from stock or seed to form a low carpet of young plants, the ‘pioneer’ phase, before increasing in height and covering the ground during a ‘building phase’. In due course, the heather achieves almost complete dominance, the ‘mature phase’, before starting to collapse, opening up the canopy and with a mosaic of dead, dying and still healthy plants; this is the ‘degenerate’ phase.

The temperature of moorland fires in each stage from pioneer to degenerate heather generally increases and becomes more variable, particularly in the degenerate phase (Hobbs et al., 1983, Hobbs and Gimingham, 1984b) and on blanket bogs and wet heaths (Currall, 1981, Hamilton, 2000). Fire intensity (heat released per unit area), will be determined primarily by the fuel load consumed, and this will normally be greater in older stands. The temperature and rate of spread of a fire will be affected by wind speed and fuel distribution (Hobbs and Gimingham, 1984b), air temperature (Brown and Davis 1973 in Allchin 1997) and relative humidity, which will influence moisture content of vegetation, litter and soil (Trollope, 1984).

On lowland heathland, the main study on fire temperature was carried out by Alchin (1997) on sites in Dorset and Devon. She generally found that fire temperatures in experimental winter fires in the vegetation were similar to those recorded in the uplands, but with lower temperatures at the litter surface than those above ground, a reversal of the situation in the uplands. Allchin (1997) concluded

51 that the fire intensities in her lowland heath fires were not high enough to remove all litter, so seed bank and rootstocks were protected from high temperatures.

The main semi-natural habitats to be affected by wildfires are moorland (blanket bog and dwarf shrub vegetation) and lowland heathland, due to the inflammable nature of the vegetation, the large areas involved which are relatively inaccessible to the fire services, and the proximity of people. Grassland, dunes, reedbeds, scrub, plantation and woodland areas can also be affected.

There is a considerable literature on the effects of controlled fires which are mostly conducted on upland moors for summaries see (Tucker, 2003, Glaves et al., 2005) but few recent studies have been carried out on the effects of wildfires. Studies on the effects of fires on moorland and blanket bog vegetation have often concentrated on controlled fires, have generally been short term, and have either been contradictory or there has been insufficient evidence to reach firm conclusions, although some general recommendations can be suggested (Stewart et al., 2004a, Stewart et al., 2004b). Most of the research into the effects of uncontrolled summer fires was carried out in the North Yorkshire Moors, after severe fires during the hot, dry summer of 1976 and in the Peak District (Legg et al., 1992, Anderson, 1986, Phillips et al., 1981, Maltby et al., 1990, Anderson, 1997, Anderson et al., 1997).

By contrast, on lowland heathland, reviews of, and research into, controlled fires is sparse but see (Alchin, 1997). There is, however, a substantial literature on wildfires on UK lowland heaths, detailing the causes of fires, the size of areas burnt and the diurnal and seasonal frequency of fires (Rose et al., 2000, Hearn and Gilbert, 1977, Webb, 1997, Tantram et al., 1999, Rose and Clarke, 2005), Other studies have examined the subsequent effects on vegetation communities, invertebrates, reptiles and birds for a review see (Underhill-Day, 2005). No studies could be found on the medium and long term effects of repeated fires (except at upland sites in Scotland), or the differences between the effects of a controlled fire in winter and a wildfire in spring or summer.

There is little information on the effects of fire on grassland habitats, and the few published studies were reviewed by Tucker (2003).

Soils Soils are protected from ‘cool’ burns by litter. Some of the fine ash particles may be washed into the soil pores, reducing infiltration rates, and increasing water retention and phosphorus levels (Mallik et al., 1984, Anderson, 1986) Water repellence on the surface of the soil may also increase in the short term after hot fires (Mallik and Rahman, 1985).

Relatively cool fires can partially sterilise the soil surface, disrupt biochemical processes increase pH slightly, and lead to nutrient loss (mostly nitrogen) through smoke and subsequent ash dispersal (Allen, 1964, Anderson, 1986). At temperatures above 300°C, nutrient loss in smoke increases rapidly (Evans and Allen, 1971)

In most cases, the main nutrients (potassium, calcium, magnesium), but not phosphorus and nitrogen, will be replaced on lowland heaths, through atmospheric inputs within a few years (Chapman et al., 1989). On calcareous grassland, chemical composition had recovered to pre-fire

52 levels within three to four years despite initial high losses of potassium and phosphorus (Lloyd, 1971). Serial burning was associated with higher manganese, potassium, magnesium, sodium and pH on moorland at Tayside, while unburned sites had higher levels of organic matter, moisture content, nitrogen and available calcium (Stevenson et al., 1996).

Disturbance by fire can also subsequently increase erosion rates through removal of vegetation and the exposure of the peat surface (Tallis, 1987, Phillips et al., 1981). In these circumstances, the soil surface is exposed to the direct effects of the sun, rain, frost and wind, and erosion will decline as the vegetation re-establishes (Tucker, 2003).

Where a severe fire has removed all vegetation and litter, or burnt into the root mat, top humus or peat layers of the soil, erosion is likely to be greater. Following severe wildfires on the North Yorkshire Moors in 1976 there was very little effect on the mineral soils and on peat soils that remained waterlogged, although the vegetation and surface litter were consumed on deeper, but drier peats, the surface was severely charred and the surviving peat formed distinctive columnar patterning, and a granular surface. This was subsequently fragmented in winter by ice crystals and freeze-thaw activity. Charred peat surfaces also broke up and were susceptible to erosion by wetting-drying, heating-cooling and freezing-thawing, producing a surface which was unsuitable for colonisation by plants. In the absence of vegetation, the authors believed these processes would continue until the entire peat mass was lost.(Maltby et al., 1990).

On blanket peats, where there are apparent sensitivities to changes in air and soil water quality, and where knowledge of biochemical processes and sediment budgets is currently limited, fires of even moderate temperature and sensitivity could present a significant risk (Glaves et al., 2005).

Where peat soils are relatively shallow, ignition can be complete and the soil reduced to a layer of ash on top of the mineral substrate. In the North Yorkshire Moor studies, nearly 30% of the ground burnt in 1976 was still bare peat in 1984(Legg et al., 1992). Another wildfire in August 1959 on Howden Moor in the South Pennines destroyed all vegetation and burnt down 1m into the peat. Ten years later, there were still extensive patches of bare peat (up to 80% of the ground in some areas) and peat erosion had been extensive with a network of gullies on some slopes, and an underground drainage system developed from cracks in the peat (Tallis, 1973). In parts of the Welsh uplands, severe fires have led to extensive loss of peat cover (Yeo, 1997).

Whilst soil erosion may be less significant on shallow soils on basic rocks, in one instance a fire destroyed the vegetation and organic, acid soil over limestone, and altered the community from acid to limestone grassland (Grime, 1963).

Vegetation The recovery of vegetation after a fire is a function of the composition of the plant community, the age of the plants since the last fire, the composition of the seed rain onto burnt areas and the characteristics of the fire itself (Hobbs and Gimingham, 1984b, Hobbs and Gimingham, 1984c, Hobbs et al., 1983, Marrs, 1986, Gloaguen, 1993) Assuming that the charred surface provides a suitable seed bed for plant re-establishment, then plants have a variety of strategies for regenerating after fire. Some produce new shoots from the base of the stem, others from buds on creeping tillers or from low leaf rosettes, or from bulbs, corms or rhizomes underground. As heather ages, there are fewer stems and flowers and the ability of the plants to re-sprout after a fire declines after about 12

53 years (Miller and J., 1970, Mohamed and Gimingham, 1970). Berdowski and Siepel (1988) found that 90% of stems sprouted from five-year old Calluna, but this figure declined to 10% in stands that were 20 years old.

Where fire kills the established plants, many species will persist through seed germination, although this takes longer initially as germination can lag behind re-sprouting regeneration by 1-2 years. Thus, burning of old stands of heather kills most of the established plants and leads to seedling regeneration, which takes longer. Fire temperatures of 500°C or more, lasting over a minute will kill the stem bases of heather, and a temperature of 200°C will kill heather seed, with short exposure to 40-60°C stimulating germination and higher temperatures and exposure times progressively reducing it (Whittaker and Gimingham, 1962, Whittaker, 1961) Most of the seed bank in heathland lies within the top 2-3 cm of soil (Legg et al., 1992).

Other moorland and heathland plants are also affected by fire temperatures, with re-sprouting generally declining with higher temperatures. Juniper is particularly sensitive to higher temperatures with little re-growth after fire at 600°C and temperatures of 800°C causing death (Mallik and Gimingham, 1985). The same authors found that seed germination in cross-leaved heath (Erica tetralix), petty whin (Genista anglica) and slender St John’s wort (Hypericum pulchrum) was stimulated by heat.

A review of a rigorous sample of studies on burning on wet heath and bog found that although results were mixed; burning tended to promote the dominance of a few species; encouraged a move from dominance of ericoids to graminoids; increased the proportion of bare ground and resulted in a decreased abundance of key species (Stewart et al., 2004a). In all cases the burnt areas were within grazing blocks and the effects of grazing after burning were not assessed. Generally there were no consistent changes in species composition following burning but a trend towards an increase in bryophytes and the promotion of dominance of Calluna vulgaris, Eriophorum spp and Molinia caerulea, The review found that there had been only one study of rotational burning and few studies examined vegetation structure (Stewart et al., 2004a).

On blanket bog, cottongrass may only gradually be replaced by heather from about 15 years after a burn (Rawes and Hobbs, 1979). However, this will depend on the age and abundance of the heather before the burn, and the heat of the fire. There are many areas where heather establishes dominance more quickly after normal winter burning on blanket bog where heather was abundant prior to burning. If heather does take time to re-establish, repeat burning at less than 20 year intervals is likely to lead to permanent dominance by cotton grasses.

A review of the effects of burning on the vegetation diversity of sub-montane dry dwarf shrub vegetation by (Stewart et al., 2004b) concluded that there was only one study which had considered rotational burning, and this suggested that the results of repeated burning depended on the stand age at the time of burning and the time that had elapsed since the last burn. The review also found that there was no significant effect of burning on heathland plant diversity overall but that the effects of a single burning cycle can reduce biodiversity in old stands. Stevenson et al (1996) found that serially burned sites on dry heath were generally more heterogeneous floristically than unburned sites, which were generally dominated by ericaceous dwarf shrubs and bryophytes. Plant species richness varied with the time elapsed since the last burn and was generally better in burnt than in un-burnt stands, except in mature stands where it was higher in the absence of burning. Too

54 frequent burning may encourage fire resistant species such as purple moor-grass and debilitate heather because it is burnt off too frequently (see Grant et al 1963; Miles 1987).

After a fire where temperature and intensity are more moderate, vegetation recovery will be largely influenced by the vegetation composition before the fire, although subsequent management, particularly grazing and trampling, will modify regeneration pathways. The less palatable or better- adapted species may be favoured by grazing, so that, for example, cross leaved-heath and the more unpalatable graminoids may benefit initially at the expense of heather. Burning of old stands of heather on a dry Scottish moorland resulted in initial colonisation by wavy hair grass (Deschampsia flexuosa) (Hobbs and Gimingham, 1984a). On wet heath, fire led to dominance by a range of graminoids that were not supplanted by dwarf shrubs for about 15 years (Currall, 1981) and on a blanket bog in the Pennines, fire led to replacement of heather by common cotton grass (Eriophorum angustifolium) for at least 15 years (Rawes and Hobbs, 1979). Where mat grass (Nardus stricta) dominance resulted from burning after an uncontrolled fire, heather had resumed dominance after nine years (King 1960). Stevenson et al (1996) found that two serially burnt stands of dry heath aged more than 19 years when burnt had lower species richness than unburnt controls.

In Dorset, comprehensive heathland surveys in 1978 and 1987 allowed an assessment to be made of vegetation recovery rates and patterns following a severe fire in 1976 (Bullock and Webb, 1995). This found that the proportions of broad habitat types present before the fire were the same 11 years later after recovery. In scrub areas, however, there was evidence of some replacement of gorse by birch, and some increase in bracken cover.

Following fires on dry heath, initial colonisation by short-lived bristle bent grass (Agrostis setacea) peaked after four years and was replaced by heather, while on wetter areas, purple moor-grass initially increased but then gave way to heather. However, on dry, humid and wet heath, neither the extent nor composition showed any effects of the fires by 1987.

On southern heathlands in Brittany, following wildfires, the initial recovery and colonisation by bryophytes and grasses can delay recovery of dwarf shrub vegetation and it may take up to 20 years for heath to return, while on the better soils, and where there is no grazing, fire may encourage the regeneration of birch woodland (Clement and Touffet 1990; Gloaguen 1990). However, occasional burning may also remove litter and slow nutrient build up, and thus help to prevent the replacement of nutrient-poor conditions dominated by ericaceous shrubs with richer, grass-dominated communities (Aerts 1990; Berendse et al 1994). However, fires to achieve this aim would be less damaging to other features if carried out intentionally and in a planned way in winter.

There is little information on rates of recovery following fire at different times of year, although heather regenerated more successfully in autumn than in spring in experimental fires in Scotland (Miller and J., 1970). Where fires are severe, causing charring and damage to the peat surface, initial re-colonisation where the soils or surfaces are stable may be by crustose lichens, which can inhibit regeneration to a heathland vegetation (Legg et al., 1992). Often colonisation is by bryophytes, possibly with a succession from smaller cushion species to the larger Polytrichum species, and can last for a decade or more (Maltby et al., 1990, Anderson, 1997).

Where fire damage has been severe, re-establishment of plants on bare peat that provides an unstable, eroding surface may be scant or absent. Without substantial management input (lime,

55 fertiliser - needed only on very acidic peat, grass nurse and heather seed), such surfaces, even with no stock grazing pressure, will remain un-vegetated for decades or longer, as demonstrated on many sites in the Peak District blanket bogs (Anderson, 1997). The dominance of purple moor grass over heather may be increased on upland wet heath by managed burning which is followed by low stocking rates (0.66 ewes/ha) (Ross et al., 2003) Repeated or hot fires are also likely to encourage bracken invasion, particularly in older stands of heather (Tucker, 2003). However, a study by (Nimes, 1995) found that in the Quantock Hills in Somerset, bracken was more likely to have replaced a dwarf shrub community if the ground had remained un-burnt during 1938-1987 than if it was burnt at least once during that period. This contrasts with (Anderson, 1986) who recorded a significantly higher rate of spread of bracken into heather moorland after a wildfire in the Peak District than in an unburnt control area.

On limestone grassland (Lloyd, 1968) found most grasses re-grew from ground level after burning, although tussocks of sheep’s fescue Festuca ovina were killed. No differences were observed in the pattern of response of perennial plants, but fairy flax Linum catharticum and some other short lived species increased initially after fire but declined again as the vegetation cover re-established. This suggested that burning may benefit species of open conditions and short turf. Lloyd also found some evidence that fire increases seedling survival of these short lived species. Hawthorn Crataegus monogyna scrub was severely damaged by fire but all the affected bushes re-sprouted.

Burning has also been found to benefit grasslands on the Culm measures in Devon, where it is generally compatible with the maintenance of species-rich grasslands, particularly where the fire is cool, takes place in winter, and is followed by cattle grazing later in the year (Wolton, 1991).

Several workers have found that fires encourage more prolific flowering in grasses, including, meadow oat grass Helictotrichon pratense, crested hair-grass Koeleria cristata, quaking grass Briza media and reed(Phragmites australis, (Lloyd, 1968, Cowie et al., 1992). Cowie and his co-workers also found that fires improved species richness and diversity in reedbeds (including increasing notable species such as marsh pea Lathyrus palustris and milk parsley Peucedanum palustre when compared to areas which had been cut or left unmanaged, but that the differences were not significant.

On peatlands, plants with growing points at or below ground can have a competitive advantage over those, such as the heathers where growing points are above ground. This can encourage the former including purple moor grass (whose shoots are above ground but protected by sheathing leaf bases), deer grass, cotton grass and purple small-reed grass Calamagrostis canescens to spread after fires. Bryophytes such as Polytrichum ssp., Aulocomnium palustre, and Campylopus introflexus can increase after fire on peatlands, helped by the release of nutrients by fire (Rowell 1988). Where tor grass Brachypodium pinnatum is present in basic grasslands, burning can encourage this coarse, aggressive grass to the detriment of other species (Crofts and Jefferson 1999).

Burning can also favour cloudberry Rubus chamaemorus and summer fires can lead to a reduction in crowberry Empetrum nigrum (Hobbs, 1984, Anderson, 1986). The effects of burning on Sphagnum spp. appear to be variable, may be less damaging where water tables are high and fires are cool, and are not always detrimental, although both mosses and lichens are likely to be damaged when fires are hot (Shaw et al., 1996, Macdonald, 2000, Tucker, 2003). Rowell (1988) reports that a fire on Glasson Moss in Cumbria virtually eliminated Sphagnum pulchrum, and that on Borth Bog the

56 number of S. imbricatum hummocks was reduced from 169 to 39 after a single fire. Cowie et al (1992) noted that bryophytes may suffer badly from burning in reedbeds.

Few studies of the effects of fire have been undertaken on non-heathland habitats. No studies of burning on sand dune communities could be found, although English Nature (2003) note that fires may damage seed banks and seedling recruitment. A study on the effects of wildfire on calcareous grassland found that annual and biennial grass species increased after burning and that invading woody species were severely checked but seldom killed outright (Lloyd, 1968). On acid grassland, large scale burning, under low intensity grazing can lead to purple moor-grass dominance and substantial areas of uniform, even-aged and species-poor habitat (Tucker, 2003).

Cowie et al (1992) noted that at one regularly burnt fen site in Broadland, burning did not appear to have any detrimental effects on the vascular flora, and had prevented colonisation by willow (Salix) and alder Alnus glutinosa scrub.

Birds There is a substantial literature on the effects of both controlled and wildfire in North America, Australia and the Mediterranean on individual species or groups, but few studies in the UK. In Wales controlled burns have historically taken place on moorland in the spring to give an early bite to sheep on the heaths and commons (‘swaling) and in more limited areas for the management of red grouse Lagopus lagopus (Yeo, 1997).

Breeding Birds

Any wildfire during the bird breeding season will normally destroy nests and eggs together with unfledged young within the burnt area, and in a severe fire, possibly adult birds which can get sucked into the fire by fierce air currents generated by the heat.

On lowland heathland only one study has looked in depth at the effects of fire on a bird species in the UK (Murison, 2007). The study looked at the effect of heathland fires on Dartford warbler Sylvia undata, a small passerine, largely confined to heathlands and strongly territorial at all times of year except in severe weather. Thus where a fire has impacted a Dartford warbler territory, the birds will continue in occupation if enough habitat has survived within or adjoining the territory to allow survival, even if breeding is not possible.

Murison found that generally Dartford warbler pairs affected by fire had fewer broods and raised fewer young overall than unaffected pairs although clutch size and nesting success was unaffected. This could partly be caused by a delay in the start of breeding as has been found for passerine species on Australian heaths where this was attributed to a lack of materials for nest building and inadequate food for egg production(Brooker and Rowley, 1991). Murison also found that Dartford warblers continued to make use of burnt territories provided some unburned patches remained. If less than half a territory was burnt then 83% of such territories would be occupied by a territorial male the next season dropping to 50% occupancy if more than half were burnt, 25% occupancy with 90% burned and absence if the whole territory was burnt. Some 45% of birds enlarged their territories to take in adjoining unburnt ground.

Overall, there was drop of 41% in Dartford warbler breeding density in sites affected by wildfires in the following year, similar to a figure recorded for Dartford warblers after fire in Mediterranean scrubland (Pons et al., 2003).

Overall, in a study of Australian heathland birds 81% of the species previously breeding were still able to nest despite the fire, mostly using only sprouting plant species as nest sites (as opposed to plants regenerating from seed), and with one species using unusually high sites in trees for two years after the fire (Brooker and Rowley, 1991) . One of these, the Splendid fairy-wren Malurus splendens suffered little direct mortality of adults due to fire, but although the wrens maintained their territories even where these had been completely burnt, post-fire breeding productivity was severely depressed due to increased predation(Rowley et al., 1991, Rowley and Brooker, 1987, Brooker and Rowley, 1991).

In Mediterranean scrub, bird abundance and species richness increased rapidly after the first year for five years after a fire with the community continuing to support open-space species as well as an increasing abundance of scrub nesting species (Herrando et al., 2002), although it has also been reported that open-space species which have colonised burnt areas can show a marked decrease in density after only two years (Pons, 1998)

The effect of fire on a grassland bird was studied by Best (1979) who investigated a population of field sparrows Spizella pusilla. She found that after the controlled burn in late April, the birds continued to establish and maintain their territories and although nest cover was severely reduced, females continued to nest in the burnt area. However birds spent more time foraging in the adjoining unburnt area. A second study of burning in grasslands showed a significant decline in numbers of breeding birds after a spring fire with 5.6 birds per km-1 of survey transect in burnt and 8.6 birds per km -1 in unburned fields, but no significant difference in nesting success between burnt and unburnt areas (Robel et al., 1998). In a study of a grassland/scrub gamebird, the grey partridge Perdix perdix it was found that the most critical effect of burning was the reduction of cover in the ground (0.05-0.25m) and shrub (0.25-0.5m) layers. After a burn the nesting and foraging habitat for chicks took 8 years to recover, but 5-6 years to provide suitable foraging for fledged young, and the authors recommended that controlled burning should not be at intervals of less than 15 years (Novoa et al., 1998).

A long term study of bird communities after a fire in Mediterranean oak forest found that bird species composition changed steadily as the vegetation recovered after the fire, with an almost complete replacement of the early successional species to late successional species over 28 years but with evidence of stabilisation towards the end of the study (Jacquet and Prodon, 2009). There is some evidence that the decline of obligate woodland species after a fire is affected by the severity of the fire and its effects on the volume of the surviving canopy the decrease in tree seeds and the reduction of nest site availability (Moneglia et al., 2009). After a forest fire, a population of Alaskan spruce grouse was reduced by some 60% the following spring, with most of the remaining birds apparently reluctant to abandon former home ranges (Ellison, 1975). It seems that species of open habitats respond to small changes in the quality of the habits with greater changes in abundance than forest species (Herrando et al., 2003)

On moorland, burning is a common practice in the uplands primarily to provide suitable young growth for sheep and red grouse. Burning may also be undertaken to promote structural diversity in

58 heather stands, to help recovery after attacks by heather beetle Lochmaea suturalis or other invertebrate or to re-invigorate old heather stands. Heather burning for grouse management tends to be in small and scattered patches with a burning cycle of 10/15 years (English Nature, 2001). Larger fires set for sheep grazing can be larger and can be haphazard and uncontrolled (Miller and Miles, 1984) with the largest burns often resulting from wildfires.

There seems to be a general agreement from a number of studies that controlled moor burning has a positive effect on densities of red grouse and golden plover and negative effects on the densities of meadow pipits (Tharme et al., 2001, Smith et al., 2001, Pearce-Higgins and Yalden, 2004, Haworth and Thompson, 1990a). The positive effects seem to be due to the provision of fresh young heather and diversification of structure for invertebrates which benefit wader and grouse chicks and adult grouse feeding on heather shoots, and the negative effects due to a decrease in habitat structure and the diversity of invertebrates although relatively little research has been done on upland invertebrates (Tucker, 2003). Moor burning may have some benefits for black grouse skylark Alauda arvensis and twite Carduelis flavirostris and be detrimental to hen harrier Circus cyaneus, merlin Falco columbarius and short-eared owl Asio flammeus if fires destroy old stands of heather in which these species nest (Smith et al., 2001, Coulson et al., 1992, Tharme et al., 2001, Backshall et al., 2001). Where taller vegetation is burnt it will usually be some time before it is again suitable for winter roosting by raptors such as hen harrier or short-eared owl. Moorland wildfires in summer will destroy the nests of ground nesting birds including passerines such as meadow pipits Anthus pratensis and skylark red grouse, raptors and waders.

These results suggest that that:

 Fires that cause substantial damage to the vegetation can lead to a severe initial decline in bird populations depending on habitat and species, but even small patches of surviving vegetation can be enough for some species in heathland and grassland to persist and breed.

 After a fire a scrub grassland mosaic, open ground species such as larks and pipits can move in and persist for some time with the recovering populations of scrub species

 Where fire damage to the ground or shrub layer vegetation is patchy, strongly territorial species will stay on their territories and can breed successfully

 After a fire, many species will stay in their territories if resources permit and will make use of remaining features for nesting where these would not normally be used in unburnt habitat

 Bird breeding in burnt areas produce fewer young than birds in unburnt areas due to later breeding and fewer broods rather then lower clutch and fledged brood sizes or lower nest success.

 Severe damage to taller scrub and tree species is likely to have a greater affect on forest species, depending on the extent of the damage and the nature and abundance of the surviving trees

 Declines in breeding bird populations could be greater in later successional stage vegetation such as woodland than in early or arrested successional stages such as heathland or grassland.

 Recovery of pre-fire levels of abundance and richness in bird populations of scrub, heathland and grassland species can be far more rapid (4-6 years) than forest species which can take 20-30 years

 In uplands red grouse and golden plover have been shown to benefit from a controlled fire regime and black grouse, skylark and twite may also benefit.

 Frequent fire could have a negative effect on meadow pipit and could deleteriously affect ground nesting raptors if old heather stands are seriously reduced in extent and distribution

Mammals

No studies on the effects of fire on mammals in the UK could be found but the general pattern of response is apparent from studies carried out elsewhere (for a review see (Clarke, 2008). Generally, the effect of a fire on ground living small mammals will be to deprive them of cover and food and make them more susceptible to predation (Green and Sanecki, 2006). As a result, small mammal populations show a marked reduction following a fire, and for some species, a decline in numbers can continue for a period of time after the fire (Zwolak and Foresman, 2007). The severity of the decline can be mediated by the presence of surviving features such a patches of grass, piles of logs, fallen timber from the fire and surviving standing live or dead wood (Stevens, 2007).

Generally, small mammals and ungulates are the first to repopulate burnt woodland, with animals such as hares returning to mid-successional woodland and bats, mustelids and squirrels returning when older growth stands have become established (Fisher and Wilkinson, 205). On moorland, it has been suggested that hares would benefit from reasonably frequent burning as this will encourage grasses for foraging, but with stands of older, denser heather for shelter (Tucker, 2003)Deer will move in after recovery of the ground vegetation in open woodland during the spring and summer after the burn, for a year or two, but after this numbers decline to pre-burn levels. The same study could detect no significant change in fawn survival rates after fire (Klinger et al., 1989). The immediate effects of a fire on survival rates and abundance in small rodent populations can be mediated by the survival of patches of vegetation after a fire in scrub communities (Monimeau et al., 2002). Depending on the severity and time of year of the fire and the type of habitat, small mammal numbers are either unaffected or rapidly recover after a fire, although populations can take longer to recover from autumn burns than spring burns (McGee, 1982). Small mammal numbers in forest are not significantly affected by fires unless these are severe (Beck and Vogl, 1972). In conifer forest, small mammal populations have been recorded as recovering to pre-fire populations after twelve months, while in grassland, the vegetation took six months to recover after which a higher abundance and diversity of small mammal species was recorded in the burnt than in unburnt habitat, although recovery could be delayed by grazing (Zwolak and Foresman, 2007).

We could find no studies of post fire effects on mammal populations in the UK, but studies elsewhere suggest that:

 Small mammal numbers recover quickly from a fire particularly where patches of ground vegetation survive the conflagration

 Larger grazing animal such as deer take advantage of the lush growth in the years following the fire, but after this their numbers fall back to pre-fire levels

 Arboreal animals such as squirrels, bats and pine martins can take many years to repopulate severely burnt woodland depending on how much live and dead timber has survived. Groups such as bats may be affected by post-fire salvage of standing dead trees

Herptiles

There is no evidence to indicate that fire is an important factor in the survival of populations of amphibians, although at times of year when animals are away from their breeding ponds, some fire induced mortality must occur. The most specialised amphibian the natterjack toad Bufo calamita, is largely confined to sand dunes and heathland habitat, where occasional fires which reduce scrub and create more open conditions may benefit populations in the long term.

There are numerous examples of the effects of fire on reptile populations, with direct mortalities as a result of fire and subsequent predation of the survivors in the aftermath with an absence of cover and food (Nature Conservancy Council, 1983). Little work has, however been carried out to estimate mortality levels and recovery rates. After a large heathland fire on an area with an estimated pre-fire population of 700-1050 smooth snakes Coronella austriaca (a species absent from Wales), only nine adults and two broods of young (five and eight) were discovered during a search after the fire, together with 30 sand lizards Lacerta agilis whose pre-fire population would been substantial (Spellerberg, 1977). It is suggested that the rate of recovery of reptile populations after fire will be affected by the extent of open habitat after fire, the speed of recovery of important plant species and the more gradual accumulation of leaf litter and dead wood (Driscol and Henderson, 2008).

Fire is seen as a major threat to reptile populations and causes the most damage during the reptile active season (March – October). Fires are particularly devastating to reptile populations because of their specific habitat requirements and restricted home ranges. Sand lizards take a particularly long time to successfully recolonise areas after a fire and can only do if so they remain nearby as it takes several years for the maturation of heather plant to provide the structural diversity in the vegetation favoured by sand lizards (Corbett & Moulton 1988 & Edgar 2002). Fires which also occur on heathland and dune habitats particularly in areas of importance for egg laying adjacent to suitable foraging habitat (Moulton and Corbett, 1999) are a concern. Depending on the intensity of the fire suitable hibernation sites for reptiles could also be lost.

Most of a population of common lizards Lacerta vivipara on an area of about 1 ha of heathland at Strensall Common in Yorkshire were killed when this was severely burnt. However, repeated surveys found that the population had recovered its estimated numbers after three breeding season, with recolonisation largely by young first year individuals coming from an unburnt adjoining heath (Simms, 1969). This account suggests recolonisation after fire may be faster in common than in sand lizard where it is suggested that the first lizards move in after about 7-10 years with first breeding after 10-15 years and restoration of the breeding population to the pre-fire size after 25 years (Nature Conservancy Council, 1983).

In summary:

 Reptiles are particularly susceptible to incineration by fire as they are slow moving and largely live above ground and have a limited home range

 Where ground cover has been removed reptiles are at an increased risk of ground and avain predation as they are largely active in daytime. With the loss of ground cover there is little or no prey availability for the surviving reptiles.

 Where patches of vegetation survive fire these can act as sanctuaries for reptiles

 In smaller fires, burnt areas will be re-colonised by reptiles from adjoining unburnt habitat within a few years, but large fires can suppress population for a considerable time or cause local extinctions particularly in a fragmented landscape.

Invertebrates The main habitats associated with high invertebrate interest and affected by controlled or wild fires are heathland, grassland (including reedbeds) and moorland, with few studies of the effects of fire on invertebrates on any of these and with those studies that have taken place often limited in scope, based on specific collection and sampling methods and without controls (Tucker, 2003). Generally the consensus from research studies is that almost all above ground invertebrates are destroyed by fire whether these are ‘hot’ or ‘cool’ fires. However, provided adjoining habitat undamaged habitat is present, then these can serve as refugia and re-colonisation can take place (Merrett, 1976, Ghandi et al., 2001, Harper et al., 2000, Tucker, 2003). Burning also reduces the populations of soil dwelling invertebrates in dry habitats (Buffington, 1966, Rickard, 1970, James, 1988, Ahlgren, 1974) but not those of wet habitat (Ditlhogo et al., 1992)

Following a fire, as the vegetation develops, the mix of invertebrates changes from early successional species of bare ground habitats to those species characteristic of the greater diversity of structure of more mature vegetation with its associated litter layer (Merrett, 1976, MV et al., 1976, Usher, 1992, Usher and Smart, 1988, Gardner and Usher, 1989, Andersen, 1991, V. et al., 1976, Brian et al., 1976). However the greatest diversity of invertebrates can be on the pioneer and degenerate phases of the heather communities on upland moors (Gimingham, 1985, Barclay-Estrup, 1974)

Recovery of the full community of unburnt areas can take as little as two years in grassland to 20 years in heathland habitats (Panzer, 2002, Bell et al., 2001) Some species and communities can benefit from the open conditions following a fire or in regularly burned sites (Johnson, 1995, Joy, 1995, Delettre, 1995, Cadbury, 1992a, Cadbury, 1992b) while others can be seriously depleted or even eliminated (Kirby, 2001) .

Where fires are extensive, whole populations of invertebrates can be destroyed and large fires may cause local extinctions in less mobile species. Invertebrate groups which are most vulnerable to fire in open habitats are those present in the litter as eggs or larvae in spring when many fires take place, species with only one generation per annum and sedentary or flightless species or groups. These include molluscs, leafhoppers, grasshoppers and some butterfly and moth species (Panzer, 2002, Kerney, 1999).

In general, burning is seen as particularly harmful to invertebrates where fires are severe or cover large areas (Kirby, 2001, RSPB, 1995, Tucker, 2003, Aked, 1984), and on particular habitats e.g. grassland (Kirby, 2001) and perhaps blanket bog (Anderson, 1986).

In summary:

 Soil invertebrates are more vulnerable to fire in dry than in wet habitats but there have been few recent studies

 In a number of habitats, especially those adapted to fire management such as heathland and moorland, invertebrate communities recover quickly from small fires

 Some invertebrate species benefit from fire management, either immediately or in the medium term

 Following fire, invertebrate communities follow a successional path related to the decline of bare ground and recovery of the vegetation, and a mozaic of habitats of different ages following fires provide a range of suitable conditions for these communities

 Large fires can be very damaging, delaying re-colonisation and threatening extinction to more vulnerable species

 Species with low vagility (dispersal abilities) or those with a life cycle vulnerable to a disturbance effect such as a fire are most at risk

 Invertebrate habitats which can be damaged by fire include grassland and blanket bog

Table 7: Seasonal habitat vulnerability issues from fires. For definitions of colours see introduction. Spring Summer Autumn Winter 1 Broadleaved, mixed and yew No literature found/relevant issues woodland No literature found. Unlikely to be a serious issue except in the uplands where slight risk of fires spreading from controlled burns out of 2 Coniferous woodland hand (and therefore not likely to be linked to recreation). 3 Boundary and linear features No literature found/relevant issues

4 Arable and horticulture No literature found/relevant issues

5 Improved grassland No literature found/relevant issues  Light fires can benefit  Light fires can benefit 6 Neutral grassland vegetation diversity if No literature found/relevant issues vegetation diversity if followed by grazing followed by grazing  Fires can encourage dominance by tor grass 7 Calcareous grassland  Fires can increase biodiversity and control woody species 8 Acid grassland  Burning can lead to reductions in biodiversity and domination by purple moor grass  9 Bracken Repeated fires encourage bracken invasion

 Repeated fires can  Wildfires can be very damaging to fauna and if severe, to  Repeated fires can encourage conversion soils, soil invertebrates and flora. encourage conversion of heathland and wet  After very severe fires with damage to organic soils can of heathland and wet heathland to grassland lead to erosion and sterility and plant communities can heathland to grassland or woodland, take decades to recover, or not recover at all without or woodland, 10 Dwarf shrub heath depending on soils and restoration measures depending on soils and grazing regimes  Widespread severe fires can reduce populations of birds, grazing regimes  Burning of old heather reptiles and invertebrates with recovery taking many  Burning of old heather stands can reduce years. stands can reduce biodiversity and affect  Fires can encourage colonisation by alien smothering biodiversity and affect the habitat of ground bryophytes the habitat of ground

Spring Summer Autumn Winter nesting raptors nesting raptors  Fires can encourage  Fires can encourage colonisation by alien colonisation by alien smothering bryophytes smothering bryophytes  Fire can increase floral  Fire can increase floral biodiversity and set biodiversity and set back scrub back scrub encroachment, but can encroachment, but can damage bryophyte  Fires can destroy nests of breeding birds, kill damage bryophyte 11 Fen, marsh and swamp communities. invertebrates and displace small mammals communities.  Invertebrate  Invertebrate communities recover communities recover quickly if unburnt quickly if unburnt habitat nearby habitat nearby  Fires on blanket bog and peatlands can reduce plant diversity and encourage graminoid dominance particularly by repeated 12 Bogs burns at short intervals (<20 years).  Hot fires can damage Sphagnum and lichen communities on bogs and lead to dominance of bryophytes mats. 13 Standing open water and No literature found/relevant issues canals 14 Rivers and streams No literature found/relevant issues

15 Montane habitats No literature found/relevant issues

16 Inland rock No literature found/relevant issues

17 Built-up areas and gardens No literature found/relevant issues

Spring Summer Autumn Winter No literature found/relevant issues 1 Saline lagoons No literature found/relevant issues 2 Coastal vegetated shingle No literature found/relevant issues 3 Rocky shores No literature found/relevant issues 4 Maritime cliffs and slopes No literature found/relevant issues 5 Saltmarsh No literature found/relevant issues 6 Mudflats and sandy beaches

7 Coastal sand dunes  Fires may damage seedbanks and seedling recruitment

Disturbance

Introduction and definitions of disturbance Disturbance can be defined as any human activity that influences an animal’s behaviour or survival. By far the majority of the literature (and there are thousands of studies), focuses on birds (for general reviews see Brawn et al., 2001, M. and J., 1999, Hockin et al., 1992, Nisbet, 2000b, Hill et al., 1997 , Sidaway and Ramblers' Association., 1990, Whitfield et al., 2008, Woodfield and Langston, 2004, Lowen et al., 2008). Disturbance does also affect mammals, herptiles and invertebrates (for general reviews see Edgar, 2002, Lowen et al., 2008, Penny Anderson Associates, 2001b). The range of studies is potentially bewildering, demonstrating a range of different impacts, in different circumstances, to different species. For example, Ficetola et al. (2007) show strong differences in the response of different species to the same source of disturbance. There is still contention about the applicability of some methods of study and the actual impacts on populations (Gill, 2007).

In this chapter we look across all taxa, but draw primarily on the avian literature as this is where by far the most has been published. We address issues such as animals fleeing due to the presence of humans in the landscape, but also impacts such as increased predation (for example because the adults when flushed betray the presence of a nest) or direct trampling (of invertebrate burrows or nests). Disturbance can also result in an increase in stress levels, reduced feeding time or the avoidance of otherwise suitable habitat. Damage to habitat structure (for example compaction of soils or reduced vegetation cover) and the consequences of such damage for particular species are addressed in Chapter 3. Direct mortality caused by fire is addressed in the fire chapter (Chapter 4).

Most disturbance events covered in this section are of course not intentional. It is important, however, to highlight that deliberate disturbance of some species is illegal. The offence of intentionally disturbing protected species occupying places used for shelter or protection was first introduced in section 9 of the Wildlife and Countryside Act 1981 (‘WCA’) and applied to species listed on Schedule 5 to the Act. A similar but slightly wider offence was introduced by the Conservation (Natural Habitats &c.) Regulations 1994 (‘the Habitats Regulations’), which prohibited deliberate disturbance of a European Protected Species wherever it occurred. Section 9 of the WCA was later amended by the Countryside and Rights of Way Act 2000 to include both intentional and reckless disturbance.

Population consequences Most studies of disturbance demonstrate behavioural effects, such as individuals changing their feeding behaviour (e.g. Burger, 1991, Fitzpatrick and Bouchez, 1998, Thomas et al., 2003, Verhulst et al., 2001) fleeing (e.g. Burger, 1998, Stalmaster and Kaiser, 1997, Blumstein et al., 2003 , Fernandez- Juricic et al., 2001, Fernandez-Juricic et al., 2005, Webb and Blumstein, 2005, Blumstein, 2003) or being more vigilant (Randler, 2006, Fernandez-Juricic and Schroeder, 2003). Other studies have focused on physiological impacts, such as demonstrating changes in the levels of stress hormones (Remage-Healey and Romero, 2000, Tempel and Gutierrez, 2003, Walker et al., Wingfield and Sapolsky, 2003) or monitoring changes in heart rate (Nimon et al., 1996, Weimerskirch et al., 2002). While behavioural and physiological studies show an impact of disturbance, it is usually difficult to understand whether the disturbance does actually have an impact on the population size of the species in question. For example, the fact that an animal flees when a person approaches is to be

67 expected and such behaviour is of course unlikely to have a major impact on the individual in question, let alone the population as a whole.

Certain impacts of disturbance are perhaps more likely to have a population impact. Direct mortality resulting from disturbance has been shown in a few circumstances (Yasue and Dearden, 2006, Liley, 1999) and many (but not all) studies have shown a reduction in reproductive success where disturbance is greater (e.g. Arroyo and Razin, 2006, Ruhlen et al., 2003, Bolduc and Guillemette, 2003, Murison, 2002). There are also many examples of otherwise suitable habitat being unused as a result of disturbance (Gill, 1996, Kaiser et al., 2006, Liley et al., 2006a, Liley and Sutherland, 2007). Very few studies have actually placed disturbance impacts in a population context, showing the actual impact of disturbance on population size (but see Liley and Sutherland, 2007, Mallord et al., 2007, Stillman et al., 2007, West et al., 2002).

In Table 8 we summarise the different impacts by which disturbance may affect a given taxa and give some examples of the kinds of species most likely to be affected.

Table 8: Impacts of disturbance and examples of species (that occur in Wales). Impact of disturbance Example of Species

Birds

Suitable habitat avoided or lower densities during breeding season Ringed plover, lapwing, black-tailed godwit, dunlin, nightjar, woodlark

Suitable foraging sites avoided pink-footed goose

Reduced intake rate of feeding birds 0ystercatcher, brent goose

Flushing, resulting in energetic costs and stress waders

Accidental dislodgement of eggs when adults flushed from nest guillemot

Direct trampling of eggs/chicks ringed plover

Chicks dying of hypothermia because adults kept away from nest herring gull, golden plover or less brooding time

Less food provided to chicks oystercatcher, marsh harrier

Increased predation of eggs/chicks nightjar, guillemot, eider

Desertion of nest puffin

Increased time spent in flight common tern

Increase in stress due to elevated metabolic rate kittiwake

Mammals

Flushing, resulting in energetic costs and stress deer spp,

Energetic costs of arousal during hibernation bat spp.

Otherwise suitable habitat avoided for breeding grey seal, otter, badgers

Suitable feeding sites avoided deer spp,

Reduced intake rate due to vigilance red deer

Change in behaviour, e.g. nocturnal foraging otter

Reduced intake rate due to vigilance red deer

Increase in adrenaline production pine marten

Disturbance of dominant male allowing young bulls access to grey seal harem

Abandonment of young harbour seal

Predation by dogs deer spp,

Impact of disturbance Example of Species

Herptiles

Direct mortality of eggs/young through trampling, horse riding, sand lizard cyclists and walkers

Loss / degradation / damage to mature structures of heather sand lizard, smooth snake plants

Displacement of egg laying females and surfacing of egg clutches sand lizard by horse riders, cyclists, walkers and dogs,

Deliberate destruction of adults adder

Collection natterjack

Predation (by dogs and cats) Natterjack, sand lizard

Invertebrates

Direct mortality, through trampling of adults and larvae tiger beetles, invertebrates associated with intertidal habitats

Flushing of individuals, with associated energetic consequences tiger beetles and other bare ground invertebrates? and changes in distribution

Disturbance and removal Invertebrates associated with deadwood

Mechanisms: the types of activity that can cause disturbance Studies have shown disturbance effects for a wide range of activities besides simply people’s presence in the landscape, for example aircraft (see Drewitt, 1999), traffic (see Reijnen et al., 1997 for a review), boats and other watercraft (Burger, 2003, Grubb et al., 2002, M. and J., 1999, Burger, 1998, Bright et al., 2003, Mikola et al., 1994), dogs (Banks and Bryant, 2007, Lord et al., 2001) and chainsaws (Tempel and Gutierrez, 2003, Delaney et al., 1999).

Some types of disturbance are clearly likely to invoke different responses. In very general terms, both distance from the source of disturbance and the scale of the disturbance (noise level, group size) will both influence the response (Beale and Monaghan, 2004b, Delaney et al., 1999). Studies that have compared different types of disturbance usually show a weaker behavioural response to:

 vehicles than people on foot (Pease et al., 2005, Rees et al., 2005, Taylor et al., 2007);

 people without dogs rather than people with dogs (Lord et al., 2001, Taylor et al., 2007, Banks and Bryant, 2007)

 slower moving sources of disturbance (Ronconi and St. Clair, 2002, Eason et al., 2009)

 smaller groups of people (Beale and Monaghan, 2005, Beale, 2007, Fernandez-Juricic et al., 2002, Taylor and Knight, 2003, Townshend and O'Connor, 1993)

 people following regular routes/trails (Fairbanks and Tullous, 2002, Pearce-Higgins et al., 2007)

There are conflicting results with studies that have compared the behavioural response when the disturbance source approaches directly compared to tangentially (Fernandez-Juricic et al., 2005, Martin et al., 2004, Papouchis et al., 2001). There is also conflicting evidence on the disturbance effects of horses. One study (Lafferty, 2002) found that birds were more likely to take flight from horse-riders compared to people on foot. In contrast Burger (1986) suggests people on horseback do not seem to threaten birds, even though horse riders frequently moved rapidly. Three studies of large mammals have compared people on foot to people on bicycles and again there is conflicting evidence. In one experimental study of chamois (Gander and Ingold, 1997) the disturbance response of chamois to joggers and mountain bikers was slightly stronger than their response to hikers. The authors suggest the faster pace of the joggers and cyclists may have posed an increased threat to the chamois. By contrast, two observational studies have either found no significant different in the distance and type of response of mammals to cyclists compared to people on foot (Taylor and Knight, 2003) or, in a study of big horn sheep in the US, sheep fled more frequently from people on foot than people on bicycles. The stronger reaction to hikers, particularly in the high-use area, was attributed to more off-trail hiking and direct approaches to the sheep.

There are also other factors which may also influence the behavioural response of animals to potential disturbance. These include habitat (e.g. Murison et al., 2007), weather (de Boer et al., 2004, Stillman et al., 2001) and the flock/group size (Borkowski, 2001, Rees et al., 2005).

Much of the variation in how animals respond to different types of disturbance can be interpreted in relation to perceived predation risk. Many authors view the behavioural response to disturbance as a trade-off between perceived risk of predation and the consequences of fleeing (e.g. lost time feeding) (Stillman and Goss-Custard, 2002a). People in the landscape are essentially predation-free predators (Beale and Monaghan, 2004b). Hence individuals that are better fed, or have a choice of sites to go to when disturbed, are more likely to flee at a greater distance, as the cost of fleeing is less. Strong behavioural responses (such as fleeing from potential disturbance at great distances) will not necessarily highlight species particularly vulnerable or sensitive to disturbance (see Gill et al., 2001b for discussion). Where a species is a quarry species and hunted then, unsurprisingly, individuals will behave differently (Bechet et al., 2003, Bechet et al., 2004) as the perceived risk of predation is higher.

Intensity of Disturbance As described above, the reaction of an animal to a potential disturbance event is dependent upon the animals’ perception of risk in the context of the cost of the response. A range of factors will potentially influence how animals respond, for example the condition of the animal, the weather, food availability etc. The intensity of disturbance (by intensity we mean the frequency at which disturbance events occur and their scale, such as the number of people) is a further consideration. These factors may influence the intensity at which disturbance may have an impact. A good

71 example of how such issues may interact comes from a French estuary system studied by Goss- Custard et al. (2006)whose models shows that oystercatchers can be disturbed up to 1.0–1.5 times per hour before their fitness is reduced in winters with good feeding conditions but only up to 0.2– 0.5 times per hour when feeding conditions are poor. This suggests that a particular intensity of disturbance may have impacts in one year but not necessarily in others.

Using their individual-based model on the Wash estuary, West et al. (2007) explored the over winter survival of a range of species (black-tailed godwit, bar-tailed godwit Limosa lapponica, Eurasian curlew Numenius arquata, Eurasian oystercatcher, red knot Calidris canutus, redshank, dunlin Calidris alba and ringed plover) in relation to disturbance, habitat loss and changes in prey abundance, and demonstrated how these factors interact. The system as a whole was predicted to be relatively insensitive to habitat loss. Black-tailed godwits were the most sensitive species, but their survival was not affected until 40% of the feeding grounds were removed. The survival of all species in the model remained high at fewer than 20 disturbances/hour. Although actual disturbance rates on the Wash were not measured during this study it is unlikely that present-day rates of disturbance represent a threat to the survival of the bird species modelled (West et al., 2002).

Assessing the intensity with which disturbance may have an impact is further complicated by habituation. Individuals will often respond differently to disturbance on busier sites (i.e. those with lots of people) than quieter sites, essentially becoming habituated to particular levels of use or activities. A number of studies have compared how animals respond to potential disturbance sources at sites with varying levels of access and have shown habituation to occur for a range of species (Bright et al., 2003, Colman et al., 2001, Enggist-Dublin and Ingold, 2003, Fitzpatrick and Bouchez, 1998, Nisbet, 2000a). Such habituation may reduce the impacts of disturbance (e.g. the amount of times adults are flushed from the nest) but may not actually totally compensate for the impacts of disturbance (Baudains and Lloyd, 2007). Not all studies have found evidence for habituation (e.g. Lord et al., 2001), and there is evidence that individuals may become habituated to some activities but not others (Enggist-Dublin and Ingold, 2003)and the issue of habituation has been identified as an important area for further study (Liley, 2007, Sutherland, 2007).

Some examples of different levels of intensity at which disturbance can have an effect are given in Table 9.

Table 9: Selected examples of levels of disturbance and consequences Species Disturbance intensity / level Effect Reference eider Visits every 3 days to breeding A significant decrease in Bolduc & colony, early and / or late in season breeding success at colonies Guillemette visited early in season (2003) oystercatcher 1 person walking across mussel bed Average time of 18 minutes for Stillman & Goss- and flushing all birds during winter an individual to return after Custard (2002b) disturbance leach’s petrel Weekly and daily handling Both reduced hatching success Blackmer et al. by 50 & 56% compared to (2004) control group ringed plover 5 visitors per day (>3km from car Nest survival 43 – 58% 3km from Pienkowski park) and over 100 visitors per day car park and 1 – 2% within 700m (1984b) (within 700m of car park) of car park. ringed plover Mean count of 70 people per 120m Beach not occupied by nesting Liley (1999) length of beach (Feb – August) birds over three year period golden plover Visitors present for more than 25% Significant decrease in Finney et al of day probability of areas of moor (2005b) being occupied by breeding birds woodlark 8 people / groups per hour through Probability of area being Mallord (2005) suitable habitat occupied less than 50% nightjar Nests within 75m of heavily used 78% failure rate compared to 0% Murison (2002) paths on lowland heaths failure rate for nests >225m from path Dartford An average of between 13 and 16 Breeding sufficiently delayed to Murison (2007) warbler people passing through a heather prevent chance of multiple territory each hour broods chough Steady rate of increase in visitor Extinction predicted within 49 Kerbiriou (2009) numbers in line with current trends years at Isle of Ouessant, France

Particular Species or Species Groups Different species will respond to disturbance in different ways (see Blumstein et al., 2005 for a review of inter-specific variation in behavioural responses to disturbance). While such variation is of interest, it reveals relatively little about a species’ vulnerability to disturbance. Species which are ‘vulnerable’ to disturbance would be those for which there is a clear population consequence of disturbance (i.e. where disturbance is likely to affect either survival or breeding success enough to impact on population size). As we have described above, there are relatively few studies that do actually directly assess population consequences of disturbance. It is however possible to make assumptions on species where disturbance is likely to have an impact and be an issue of concern to conservationists:

 Species where disturbance has been shown to have a clear effect on breeding success or survival;

 Species where disturbance has been shown to cause individuals to not use (or markedly under-use) considerable areas of otherwise suitable habitat;

 Species with particular aspects to their ecology that might make them particularly vulnerable, even where there is a lack of direct study on the species in question.

In general terms there is also likely to be concern with respect to vary rare species. In the following sections we highlight species that occur in Wales and which we consider potentially vulnerable to disturbance.

Mammals When compared with the wealth of research relating to human disturbance of birds, there are few studies of mammals in the UK and disturbance. A number of terrestrial mammals are protected from intentional or reckless disturbance under the Wildlife & Countryside Act 1981 (‘Schedule 5 species’) and / or the Habitats Regulations 1994 (‘European Protected Species’), including all species of bats, common dormouse, otter, pine marten, red squirrel. A change to the Wildlife and Countryside Act in April 2008 now makes it an offence to disturb water voles, whereas previous protection only extended to protection when in their burrows. This legal protection provides a guide for some mammal species for which disturbance may be important. We discuss these species (and some additional ones for which there is evidence of disturbance impacts) below:

Seals

There is a wide range of studies addressing disturbance to pinnipeds, but relatively little from the UK (see Lowen et al., 2008 for review). The studies show a range of behavioural responses, with most studies focusing on hauled-out seals fleeing a source of disturbance. For some species such as the Mediterranean and Hawaiian monk seals, human disturbance is clearly a key factor in their population decline. Brown and Prior (1998) studied Harbour Seals on Mousa SAC in the Shetlands, looking at the frequency with which hauled-out seals were seen to flee when disturbed and enter the water. The authors’ primary concern was for mothers and pups during June and July. Human interference caused mothers to abandon pups, or to abandon ideal nursery sites. Prolonged disturbance caused seals to abandon haul-out sites. They suggest that some form of visitor control is necessary to reduce disturbance.

Research on grey seals at Donna Nook in Lincolnshire (Lidgard, 1996, in Saunders et al 2000) found that females preferred to give birth in areas of low disturbance and that pups born in such areas gained weight more quickly than pups born in areas with greater disturbance. During periods of high human disturbance, females were more protective towards their pups and the pups were more vigilant. While able to show an apparent impact of disturbance, the study does not identify any population consequences of human disturbance. The colony has dramatically increased in size since 1990 and the weaning and growth rate of pups was higher than those reported in other colonies.

Work on grey seals in Wales indicates that disturbance is site-specific, with seal behaviour varying between sites in how seals respond to disturbance (Westcott and Stringell, 2003). Westcott and Stringell also suggest that there is variation between individuals in how they respond to disturbance, except when in groups, where extreme reaction by one individual tends to be followed immediately by the group. This has implications for how each site and any disturbance should be managed, implying that site-specific measures may be necessary to minimize disturbance. Westcott and

Stringell also suggest that disturbance levels at the sites studied have probably varied over time and that seals appear to have become habituated to levels of disturbance that, in the 1980s, they would probably not have tolerated.

In Norfolk, monitoring results highlight the recent increase in numbers of grey seals along the East Anglian coastline and the scarcity of harbour seals (Skeate and Perrow, 2008). The authors suggest that disease has precipitated a decline from which harbour seals seem unable to recover; harbour seals now seem unable to breed on the mainland, “presumably because of the pressure of humans and their dogs”. This species may therefore be particularly vulnerable to disturbance impacts, whereas the grey seal, favouring rocky outcrops and offshore islands may be less vulnerable. Harbour seals are rare in Wales but breeding has been reported in the vicinity of Prestatyn and of breeding and Ynys Dulas had a regular group of 4 common seals hauled out during a survey in 2001 (records summarised in Lerwill et al., 2003).

Bats

Excessive disturbance is likely to cause bats to abandon a site or possibly be a cause of mortality, but there is little documented evidence or examples, probably due to the level of legal protection afforded, which would prevent a large number of experimental studies. Disturbance to bats by direct handling is relatively well documented (e.g. Steadman, Webb and Racey 1991). However there is also increasing evidence that disturbance without contact can also have a negative effect upon bats and their reproductive success. Disturbance to bats during their hibernation period can be particularly harmful, with light, temperature changes and noise all causing bats to be aroused from their hibernation. This causes energy loss during a time when conserving fat stores is critical. Tomas (1995) found that human visits to a bat hibernaculum caused bats to become active and fly, even without any physical contact from the human visitors. Maternity roosts are also particularly sensitive to disturbance, and with bats only raising one young per year, the consequences of maternity roost disturbance on population levels can be significant (Mann, Steidl and Dalton 2002).

The Colorado Bat Conservation Plan (Colorado Bat Working Group 2007) considers the impacts of disturbance to cave dwelling bats, citing torch light, talking and even people simply moving through a cave as activities that can cause the bats to temporarily move or even permanently abandon a cave, or cause females to abandon their young.

Disturbance to bats can also occur when they are actively foraging. Lighting can cause disturbance to bats and has been shown to disrupt flight patterns (Stone et al., 2009).

Otters Lutra lutra

While otters have been shown to tend to avoid areas with high levels of human disturbance (Prenda et al., 2001, Robitaille and Laurence, 2002), they are also present in many towns and cities within the UK and therefore it is thought that they are able to habituate to disturbance (Chanin, 2003). Features of breeding sites and dens are summarised by Liles (Liles, 2003), who highlights that otter holts tend to occur in areas with little or no disturbance, but that this is certainly not always the

75 case. It therefore seems that there is little strong evidence that disturbance is a real issue for this species. McCafferty (2005) describes disturbance to otters from anglers, walkers and dogs as a short term influence on the behaviour of otters, however he acknowledges that the longer term effects of these types of disturbance on breeding females and young is less known. Dogs are cited as a cause of disturbance by Liles where natal dens are in above-ground cover such as scrub and reed beds.

Water voles Arvicola terrestris,

The water vole is a riparian species that favours rivers with well vegetated banks and emergent vegetation (Harris and Yalden, 2008). The species was once widespread but has undergone rapid decline in the UK over the last 90 years or so. The two factors largely responsible for this decline are likely to be changes in habitat and predation due to the feral American mink Mustela vision. There are no studies known to the authors showing that recreational disturbance does have any impact on this species. The recent changes to the legislation to make disturbance to water voles a criminal offence, irrespective of whether in their burrows or not, primarily relates to the need to protect water voles from development and inappropriate maintenance of banks adjacent to rivers and other water bodies utilised by water voles.

Deer

There are a number of different deer species present in Wales, including the red deer Cervus elephus, roe deer Capreolus capreoulus and the introduced fallow deer Dama dama, muntjac deer Munitacus reevesi and Sika deer Cervus nippon. Deer respond to disturbance and there is evidence that their distribution, physiology, breeding success and survival can all be affected by disturbance (e.g. Mitchell et al., 1977, Yalden, 1990). Domestic dogs can chase (and sometimes kill/main) deer (see Taylor et al., 2005 for review). As deer are generally increasing in the UK and most species are introduced, the nature conservation impacts are potentially limited.

Badgers Meles meles

Badgers tend to emerge later and change their behaviour when leaving the sett if disturbed (Neal, 1977) and in a study in Italy, disturbance was the second most important factor affecting badger sett-site selection (Prigioni and Deflorian, 2005). Badgers are relatively widespread in Wales, occur across a range of habitats and there is no evidence that disturbance has any population consequences.

Pine marten Martes martes

The pine marten had become extinct throughout much of Britain by the early part of the 20th century. Small populations survived in Wales but the species remains exceptionally rare in the UK outside of Scotland. In Wales the main strongholds are Snowdonia, the Cambrian Mountains and

76 the uplands of central southern Wales. There is only one study of this species and disturbance. Barja et al. (2007) recorded adrenal activity in European pine marten, and found a physiological stress response closely related to human disturbance. They also found that this response was higher during the pine marten breeding season.

Red squirrel Sciurus vulgaris

The main sites for red squirrels in Wales include the central region (Tywi/Irfon/Crychan forest complex); Clocaenog in North Wales and Anglesey. These areas are all primarily commercial woodland plantations partially or fully owned and managed by Forestry Commission Wales9. The species is declining rapidly and is a BAP priority species. There is guidance for forestry staff that warns that recreational access or noisy activities may disturb this species, especially during the breeding season10, however experimental disturbance of a similar species in the US has shown no effect of disturbance on squirrel abundance (Gutzwiller and Riffell, 2008) and disturbance is not cited as a threat to this species in it’s UK BAP11. Although not directly correlated with human disturbance, Randler (2006) found that red squirrels will respond to alarm calls by jays, noting that the squirrels would spent less time in the area, expressed a higher vigilance, and also displayed more rapid head and body movements.

Birds The enormous volume of literature on disturbance to birds means that there is evidence for disturbance effects for a wide range of species yet few studies place the disturbance in any kind of population context or identify whether disturbance is actually a conservation issue. In order to highlight particular species or habitats we have reviewed all red list and amber species that occur in Wales (see Appendix 1). We have also referred to guidance from England (see Appendix 1 and Lowen et al., 2008). From this list we can highlight the following:

 Winter raptor roosts (e.g. harriers)

 Breeding raptors (e.g. hen harrier, peregrine, red kite) and owls (short-eared owl, Barn owl)

 Geese and swans wintering on grassland/arable sites

 Colonial nesting seabirds: gulls, terns, puffin, cliff-nesting species

 Black grouse leks

 Upland breeding waders

 Breeding waders on beaches

9 Wales Squirrel Forum: http://www.snh.org.uk/ukredsquirrelgroup/WalesForum.asp

10 http://www.forestry.gov.uk/pdf/Guidancenote33Redsquirrel.pdf/$FILE/Guidancenote33Redsquirrel.pdf

11 http://www.ukbap.org.uk/UKPlans.aspx?ID=565#2

 Breeding waders inwet grassland

 Wintering waders and waterfowl on estuaries and sandflats

 Heathland breeding species: nightjar, woodlark and Dartford warbler

 Ground nesting passerines such as skylark

 Chough, particularly feeding flocks post breeding

Herptiles Edgar (2002) identifies the sand lizard as by far the UK’s herptile most vulnerable to the effects of public access, due to its specific habitat requirements and biology. Sand lizards are most frequently found clustered at particular areas of suitable habitats, which comprise of areas of open ground within or adjacent to structurally diverse mosaics of vegetation . Edgar considers that such specific habitat requirements, clustered populations, plus their behaviour (they spend more time basking and are particularly site faithful due to their limited home range) means that they are particularly vulnerable to disturbance. In addition, the sand lizard is the only British reptile to deposit egg clutches in open areas of bare ground, excavating a burrow with a large cavity to lay between four and twelve eggs. The areas of sandy substrata selected by females are unshaded and receive large amounts of sunlight and, on many sites, these areas are only present along paths and tracks. The nests, 4–10 cm below ground level are believed to withstand light trampling by people, but are highly vulnerable to and can be surfaced by heavy trampling, from vehicles, bicycles or horses.

Edgar ranks the adder as the second most vulnerable herptile (after the sand lizard) to impacts from recreation. He draws on anecdotal evidence to suggest that adders can be surprisingly insensitive to the presence of humans. Disturbance can result in adders being flushed from basking locations and can force them to expend energy in cold weather. Edgar also documents evidence for direct mortality as a consequence of people’s fear of a poisonous snake.

The only amphibian for which disturbance can have an impact is the natterjack, a species which is limited to one site in Wales where it has been introduced. In his review, Edgar suggests that the species is relatively unaffected by public access pressures and indeed may even benefit from some otherwise undesirable impacts such as the heavy trampling of vegetation. Disturbance can however occur as natterjacks prefer to spawn in temporary ponds and the shallow water means their spawn is more easily damaged, especially by dogs running through the ponds, and the tadpoles more readily collected.

Invertebrates For many invertebrate species it is difficult to untangle the importance of access levels in modifying habitats (discussed in the damage section) from the direct impacts of trampling and disturbance. Disturbance and direct mortality from trampling are however clearly issues for some species or particular groups and we highlight these below:

Belted Beauty Lycia zonaria

Access issues relating to this BAP priority species are discussed in (Lowen et al., 2008). At Morfa Conwy the moth faces imminent extinction primarily because of the loss of semi-fixed dune grassland to succession (Howe et al., 2004). While trampling can help to maintain the required habitat, the core population is now found on an area open to high human pressure (from walkers, dogs and picnickers), where it appears that the larvae and adults are very vulnerable to trampling. In this example, succession has caused a change in distribution of the species, and access has increasingly become an issue for the species.

Deadwood invertebrates

On sites with public access, bark stripping and breaking up rotten wood can do considerable damage to the invertebrate fauna (Kirby, 2001). Removal of deadwood and dead and rotting trees is a major threat to saproxylic (deadwood) invertebrates (Alexander et al., 2005f, Alexander et al., 2005g, Alexander et al., 2005h, Alexander et al., 2005i, Kirby, 2001). This micro-habitat is often removed for health and safety reasons and also by the public for camp fires etc. Anderson (1992) discusses the consequences of the removal of deadwood for camp fires for woodland invertebrates.

Species associated with trackways and paths

Access has been shown to impact on behaviour for tiger beetles in Turkey (Arndt et al., 2005). Arndt et al. looked at three different dune areas with varying levels of human activity. The activity of adult tiger beetles diverged during the tourist season, decreasing markedly with increased disturbance. Larval activities showed similar trends, with first and second instar larvae practically absent from the heavily disturbed section. For diurnal, surface-feeding species that occur on bare ground(such as paths and tracks) disturbance can clearly be an issue. Trampling may also result in direct mortality. Observations of C. hybrida in France noted a high mortality of adults on a well-used cycle track across dunes, due to crushing by bike tyres b(Alexander et al., 2005b). Liddle noted 10 times fewer invertebrates and a much smaller number of species on trampled dunes (with only 50 tramples per week) compared with adjacent untrampled areas. Anderson (1992) discusses the implications of trampling in woodland habitats upon soils, with a decrease in soil nitrifying bacteria and effects upon soil dwelling invertebrates.

Bayfield (1979) undertook a specific study on the effects of trampling on Molophilus ater, a peatland cranefly, finding significant effects upon the species, mainly due to physical crushing having the potential to kill a high proportion of larvae within the peat. Lower numbers of larvae were found by Bayfield along pathways as a consequence.

Rocky shore invertebrates

Various studies have found trampling damage to rocky shore fauna, with both macrofauna (some bivalves, anemones, barnacles, limpets, whelks, sea stars, amphipods, polychaetes, isopods, oligochaetes and gammarids) and meiofauna (nemotodes, ostracods, acarids, tanaids, some bivalves, polychaete, oligocheate and annelid worms, sponges and caprellid amphipods) reduced in number in trampled areas (Silbernagel, 2008, Casu et al., 2006b). Some species are immediately affected by trampling and show rapid declines; these include some nematode worms, mites, bivalves, gammarid shrimps, sea urchins, isopods and polychaete worms (Casu et al., 2006a).

Sabellaria alveolata polychaete worms create biogenic reefs recognised as a UK BAP priority habitat. The reefs occur subtidally in the Severn but on the west coast of Wales they can colonise hard substrates such as boulders and outcrops on sandy beaches where they can occur well within the intertidal. At low tide the exposed reefs are highly susceptible to persistent trampling which, when combined with damaging wave action, can result in significant erosion and paths forming through the reef (Cunningham et al., 1984, Holt et al., 1998, Tyler-Walters and Arnold, 2008b).

Some studies have found that polychaete worms were particularly sensitive to trampling effects and took longer to recover than other groups (Brown and Taylor, 1999, Casu et al., 2006b). Prescott (2006) found a reduction in the presence of anemones and the amount of barnacle cover in intertidal habitats where human visitation was high.

Not all studies have demonstrated effects of disturbance on barnacles, limpets and whelks (Jenkins et al., 2002, Beauchamp and Gowing, 2003), but heavy trampling reduced the density of mussels (Van de Werfhorst and Pearse, 2007, Smith et al., 2008). In one long-term study, the density of all species studied (except small gastropods) declined over time (Addessi, 1994).

Damage to communities tends to be greatest closer to main access points (Addessi, 1994). The damage may be less where people are barefoot, and one study found 85% of those visiting the rocky shore used no footwear (Bally and Griffiths, 1989).

There seem to have been few studies on the effects of harvesting on rocky shores. The main species are shellfish, crustacea and sea urchins for bait or food, and the general effect of over-harvesting is to reduce populations and remove old adults. As most of the collected species are broadcast spawners, relying on the release of huge numbers of eggs or larvae, the population relies on a high number of fertile individuals to optimise reproductive effort, and unwise harvesting can therefore seriously compromise this (Murray et al., 1999). Harvesting can also change the relationships of predator and prey, leading to a decline in predator numbers, or to over predation of a reduced population and a switch to dominance by other species. Where human harvesting of a muricid gastropod predator of mussel beds was stopped, there was a switch to dominance by barnacles and an increase in species diversity as predation of the mussel beds increased (Duran and Castilla, 1989).

Species associated with Mudflats

In a study of trampling effects on mudflats in the Netherlands, trampling had clear impacts on the macroinvertebrate fauna (Rossi et al., 2007). Trampling resulted in reduced abundance of the adults of a clam species, the Baltic macoma Macoma balthica and cockles Cerastoderma edule. It was believed that footsteps directly killed or buried the animals, provoking asphyxia. Conversely, for the clam trampling indirectly enhanced the rate of recruitment, while small-sized cockles did not react to the trampling. The number of small animals showed little change because trampling occurred during the growing season and there was a continuous supply of larvae and juveniles. In addition, trampling might have weakened negative adult-juvenile interactions between adult cockles and juvenile clams, thus facilitating the recruitment. This work suggests that during the growing season recovery can be fast, but in the long-term it might lead towards a shift in community dynamics, possibly affecting ecosystem functioning.

Work on nematodes in mudflats has shown nematode abundance and species composition to be reduced on trampled plots, but that recovery is rapid (36 hours), suggesting that the nematodes respond to the trampling by burrowing more deeply and soon return (Johnson et al., 2007).

The effects of varying intensities of human trampling on sandy beach macrofauna were investigated experimentally in South Africa (Moffett et al., 1998). Vigorous beach games (volleyball) resulted in damage to the four species (two mussels, a mysid shrimp and an isopod) studied. One of the mussel species was particularly vulnerable to the trampling, with 18% of individuals damaged in the treatment with the highest intensity of trampling. The results indicated that few members of the macrofauna were damaged at low trampling intensities but substantial damage occurred under intense trampling.

Sandhoppers on lower beaches have been a focus for a number of studies. Declines in densities of sandhoppers on sandy beaches in Poland have been attributed to tourist pressure and the number of people on the lower beaches (Weslawski J.M. et al., 2000), with similar results in Spain and Brazil (Veloso et al., 2008). Veloso found higher densities sandhopper in protected areas where access was controlled, in both countries. Different species were involved in Spain and Brazil and the beaches differed in the variation in access through the year, yet impacts of access were found at both locations. Ugolini et al (2008) found a negative correlation between sandhopper abundance and the number of people at given locations. The people were counted from very specific areas (within 150 m of the sandhopper sampling locations) across a range of different beaches. They also included a number of other variables, included substrate size and trace metals in their analysis and conducted experimental trampling of sandhoppers. Their work shows a clear and very strong trampling effect for this invertebrate group. Some evidence from Spain demonstrates differences in the morphology of sandhoppers on busier beaches (Barca-Bravo et al., 2008), with greater asymmetry among sandhoppers at the site with the most tourist and urban pressure.

Seasonality and temporal variation Disturbance has different impacts at different times of year for all taxa. During the breeding season individuals will usually be tied to specific locations and young may be particularly vulnerable to trampling or predation. During the winter individuals will be free of the constraints associated with breeding but disturbance issues may interact with the weather, and the energetic costs of disturbance may be more severe in cold weather, particularly for groups such as shorebirds or reptiles.

The bird breeding season ranges from January for species such as crossbill through to August for some species (such as nightjar or ringed plover). For mammals there can be similar variations, for example Harbour seals’ breeding and moulting season lasts from June to August whereas badgers give birth between mid December and April). It is species for which the breeding season coincides with the ‘peak’ tourist and recreational season (i.e. July and August) that disturbance may be a particular issue, especially those that occur in habitats such as beaches that tend to be particularly busy in the summer. .

During the non-breeding season, the main impacts of human disturbance on birds is interruption to foraging and, to a lesser extent, roosting (Woodfield and Langston, 2004). There is a body of

81 research suggesting that responsiveness to disturbance is a species-specific trait (e.g. Yasué, 2005). The extent to which disturbance affects the actual distribution of individuals within a site will vary according to the species involved, the availability of other resources and the individuals’ own state. If under stress, for example during cold winter weather when food resources are scarce, birds may be less easily disturbed than at other times (Burton, 2007, Stillman and Goss-Custard, 2002a), they may simply not be able to afford to stop feeding.

There may also be seasonal variation within a species’ responsiveness to disturbance, as individuals alter their threshold in response to shifts in the basic trade-off between increased perceived predation risk (tolerating disturbance) and the increased starvation risk of not feeding, i.e. avoiding disturbance (Stillman and Goss-Custard, 2002a). Towards the end of winter, migratory birds need to increase feeding rates to provide energy for migration to breeding grounds. As winter progresses, Eurasian oystercatcher energy requirements increase and their feeding conditions deteriorate. To survive they spend longer feeding and so have less spare time in which to compensate for disturbance. Later in winter, birds approach a disturbance source more closely and return more quickly after a disturbance. Their behavioural response to disturbance is less when they are having more difficulty surviving and hence their starvation risk (avoiding disturbance) is greater (Stillman and Goss-Custard, 2002a). It is thus important to measure subtle behavioural changes in foraging rates along with key ecological variables in order to assess the true impact of human disturbance on migratory shorebirds (Yasué, 2005).

Particular habitat issues Seasonal impacts by habitat are summarised in Table 10. There are some particular habitat issues which we summarise below:

Coastal habitats seem particularly critical, with issues relating to impacts from access to seals, breeding seabirds, breeding waders, wintering waterfowl, herptiles and invertebrates. Access on the coast is often concentrated in a narrow strip that is also important to wildlife, offering little physical space for segregation of wildlife and people (Woodfield and Langston, 2004).

Sand dunes are the only habitat in Wales that supports the potentially vulnerable herptiles (sand lizard and natterjack). Both species have been reintroduced to Wales and are very limited in their distribution (for example natterjacks occur at just one site at Talacre, near the Dee Estuary).

There are a suite of breeding bird species associated with lowland heathland – nightjar, woodlark and Dartford warbler – for which there is a strong body of evidence for disturbance effects. Heathlands are also important for a number of invertebrates for which disturbance has an impact, and in winter heathland sites may be used by roosting raptors.

Wintering wildfowl such as geese and swans may use a range of different habitats, for example arable fields. Such feeding grounds tend to be relatively predictable and occur close to coastal sites / estuaries used for roosting.

A number of upland bird species may be found across a range of habitats and are therefore not necessarily captured well in the summary table (Table 10). For example black grouse may occur in coniferous woodland, grassland or heathland habitats.

Table 10: Summary of seasonal impacts from disturbance, by habitat. For definitions of colours see introduction. Spring Summer Autumn Winter 1 Broadleaved, mixed and yew  Maternity roosts of bats vulnerable to disturbance  Bat hibernation sites woodland Deadwood invertebrates  Ground-nesting birds  Ground-nesting birds (nightjar and (nightjar and woodlark) woodlark) 2 Coniferous woodland No literature found/relevant issues  Breeding raptors (e.g.  Breeding raptors (e.g. goshawk) possibly goshawk) possibly vulnerable vulnerable 3 Boundary and linear features No literature found/relevant issues  Wintering waterfowl (particularly geese and 4 Arable and horticulture No literature found/relevant issues swans), on large fields near coastal / estuary sites.  Wintering waterfowl  Ground-nesting passerines such as skylark No literature 5 Improved grassland (particularly geese,  Breeding waders in coastal grazing marsh found/relevant issues swans, duck).  No literature found/relevant issues 6 Neutral grassland Ground-nesting birds (e.g. skylark)l

 No literature found/relevant issues 7 Calcareous grassland Ground-nesting birds (e.g. skylark)l  8 Acid grassland Ground-nesting birds (e.g. nightjar, skylark) No literature found/relevant issues No literature found/relevant issues 9 Bracken  Ground-nesting birds (e.g. nightjar, skylark)  Ground-nesting/low  Ground-nesting/low  Winter raptor roosts 10 Dwarf shrub heath  Winter raptor roosts nesting birds: Dartford nesting birds: Dartford  Adder (emerges in

Spring Summer Autumn Winter warbler, woodlark, warbler, woodlark, February) nightjar nightjar  Breeding raptors (e.g.  Breeding raptors (e.g. merlin, hen harrier) merlin, hen harrier)  Invertebrates  Black grouse leks associated with paths  Invertebrates associated with paths  Otter holts / dens No literature found/relevant  Winter raptor roosts possibly vulnerable  Breeding waders 11 Fen, marsh and swamp issues  Wintering waterfowl  Breeding waders  Bittern (geese, swans, duck)  Bittern No literature found/relevant  Winter raptor roosts 12 Bogs  Breeding waders  Breeding waders issues  Wintering waterfowl  Wintering waterfowl  Wintering waterfowl 13 Standing open water and  Breeding waterfowl  Breeding waterfowl (particularly geese, (particularly geese, canals swans, duck) swans, duck)  Otter holts / dens 14 Rivers and streams No particular issues  Pine Marten?  Pine Marten? possibly vulnerable No literature found/relevant issues 15 Montane habitats  Pine marten?  Pine Marten?

 Breeding raptors (e.g.  Breeding raptors (e.g. No literature found/relevant issues 16 Inland rock peregrine) peregrine) 17 Built-up areas and gardens Maternity roosts of bats vulnerable to disturbance Bat hibernation sites

Spring Summer Autumn Winter

No literature found/relevant issues  Wintering waterfowl (geese, swans, duck and 1 Saline lagoons waders)

No literature found/relevant issues 2 Coastal vegetated shingle  Ground-nesting birds  Gull colonies  Wintering waders  Invertebrates  Invertebrates 3 Rocky shores  Invertebrates  Invertebrates (trampling damage) (trampling damage) (trampling damage) (trampling damage)  Breeding seabirds and gull colonies  Breeding seabirds and  Chough (though late  Adder (emerges in 4 Maritime cliffs and slopes  Chough (though late gull colonies summer probably key February) summer prob key  Chough time) time)  Breeding waders (e.g.  Breeding waders (e.g. redshank) redshank) 5 Saltmarsh  Gull colonies  Wintering waterfowl (geese, swans, duck and waders)  Gull colonies  Passage waders and

waterfowl  Breeding waders  Breeding waders (sandy beaches (sandy beaches)  Wintering waterfowl (geese, swans, duck and waders) 6 Mudflats and sandy beaches  Harbour seals  Trampling can damage Trampling can damage invertebrate fauna  Trampling can damage invertebrate fauna invertebrate fauna  Breeding waders  Tern colonies No literature found/relevant issues 7 Coastal sand dunes  Natterjack (one site in Wales)  Sand lizard  Belted Beauty (adult/caterpillars present March – July)

Summaries by habitat In this section we summarise the review section by habitat, outlining the impacts from recreation for each habitat.

Broadleaved, mixed and yew woodland  Pathogens spread by humans and dogs can affect a wide range of habitats. Phytophthora ramorum and kernoviae are known to affect around 40 different species of trees and shrubs.  Small mammal entrapment in human litter occurs all year and individual pieces of litter can persist for many years.  Lower plant damage can be caused by repeated dog urination.  Woodlands typically have a ground flora that is adapted to shade, and therefore plant species that have large leaves with thin cell walls. This renders woodland plants less able to withstand abrasion. Plant abrasion is particularly harmful in the late spring and early summer when growth and food storage is most critical. Trampling of woodland flora therefore a particular impact. Specific recreational activities such as paintballing in woodland can lead to drastic loss of ground flora.  Wet woodland ground flora can be lost through continued trampling.  An increase in recreation in older woodlands with less robust veteran trees can lead to increased tree death, particularly associated with increased trampling around the tree roots. Compaction in Autumn months can affect soil fungi, and this in turn affects the health of trees.  Woodland fires can have detrimental effects on arboreal woodland mammals, possibly due to the reduction in standing deadwood in addition to direct mortality.  Loss of deadwood can occur on sites with high recreational use where there is pressure to ensure the safety of people. This reduces habitat for invertebrate deadwood specialists.  Disturbance such as noise and lighting around trees can adversely affect tree dwelling mammals. Winter hibernation sites for mammals can be disturbed by noise and lighting, and an animal aroused from hibernation is at risk of predation, exposure to the elements or starvation.

Coniferous woodland  Lower plant communities typically found in coniferous woodlands have been shown to suffer long term damage and loss from trampling, even from single recreation events.  Nightjar and woodlark may nest on the ground in clearfell areas. Areas of high disturbance likely to be avoided and breeding success also potentially affected by disturbance.  Disturbance may be an issue for breeding raptors (goshawk).  Whilst there is little evidence to indicate that human disturbance has a detrimental effect upon red squirrels, the wider vulnerability of this species suggests that risks should not be taken with regard to high levels of human disturbance in sites where red squirrel populations remain.

Boundary and linear features  Linear features such as hedgerows are used by bats to navigate around the countryside and forage on flying insects. Disturbance-related impacts upon boundary features, such as noise and lighting can affect bat flight paths and prevent access to foraging sites. This is particularly relevant during the active time of year for bats, from spring to autumn.

Arable and horticulture  Waterfowl such as geese and swans can gather on arable fields in some areas (typically close to estuaries or the coast). Disturbance may result in areas being avoided and increased energetic costs for the birds.

Improved grassland  Grasslands, and particularly improved grasslands, are significantly more resilient to trampling than other more fragile habitats. Improved grasslands already support a range of species that are semi-tolerant of trampling. However, heavily used sites will gradually go through a change in species composition towards greater densities of the most tolerant species.  Invertebrate communities can be directly affected by trampling, irrespective of grassland type or nutrient levels. Trampling can reduce the litter layer and cause a decline in the numbers and species of litter fauna.  Little work has been undertaken on the effects of grassland fires, but clearly grasslands are most vulnerable to fire at the driest times of the year during the summer.

Neutral grassland  Nutrient enrichment, particularly from dog fouling, is an issue and is generally concentrated around car-parks and access points and often along path edges. This is an issue all year round.  Unmanaged grass, i.e. grass which has not been the subject of regular cutting or grazing, tends to be the most susceptible to damage from trampling, not having built up a tolerance to regular disturbance.  Little work has been undertaken on the effects of grassland fires. Grasslands are most vulnerable to fire at the driest times of the year i.e. during the summer. Light fire damage can be beneficial and followed by an increase in plant diversity.

Calcareous grassland  Nutrient enrichment, particularly from dog fouling, is an issue and is generally concentrated around car-parks and access points and often along path edges. This is an issue all year round .  Heavy trampling can alter soil chemistry, by affecting soil structure, oxygen and water content, and levels and types of bacteria. This can be damaging to more alkaline habitats, where soil pH is reduced to a more neutral level.

 Poor nutrient levels typically found in calcareous grasslands result in slower plant productivity, meaning plants are less able to recover from trampling damage.  A number of calcareous grassland invertebrates have a preference for patches of bare ground, and some bees and wasps require these small habitat patches for burrowing. Bare patches created by light disturbance are therefore beneficial. Heavier levels of disturbance and the addition of horses, cycles and motorcycles may be detrimental as the levels of erosion and compaction make bare ground unsuitable for use by invertebrates.  Little work has been undertaken on the effects of grassland fires, but clearly grasslands are most vulnerable to fire at the driest times of the year during the summer. Studies on the recovery rate of calcareous grassland have shown a return to pre-fire chemical composition within three or four years. Other studies indicate beneficial consequences of fire for calcareous grassland, with opportunities for different species to re-colonise for a few years until the full sward redevelops.

Acid grassland  Nutrient enrichment, particularly from dog fouling, is an issue and is generally concentrated around car-parks and access points and often along path edges. This is an issue all year round  Heavy trampling can alter soil chemistry, by affecting soil structure, oxygen and water content, and levels and types of bacteria. Compaction from trampling can effect earthworm, and other soil arthropod populations. This can be damaging to more acidic habitats, where soil pH is increased to a more neutral level.  Poor nutrient levels typically found in acidic grasslands result in slower plant productivity, leaving plants less able to recover from trampling damage.  Little work has been undertaken on the effects of grassland fires, but clearly grasslands are most vulnerable to fire at the driest times of the year during the summer. Fires within acidic grassland habitats can be followed by an over-dominance of purple moor-grass, thus reducing species diversity.

Bracken  Bracken is a robust and relatively damage tolerant plant. The plant is most vulnerable in early stages of growth.  Continued trampling can result in increased frost damage to rhizomes.  Dry stands of bracken with significant litter layers are particularly susceptible to fire during the summer months.  Bracken can benefit from fire damage, where opportunities are created for bracken to spread, for example on moorlands.

Dwarf shrub heath  Nutrient enrichment, particularly from dog fouling, is an issue and is generally concentrated around car-parks and access points and often along path edges. This is an issue all year round. Certain restricted plant and lichen species such as pillwort Pilularia globulifera, pale

dog-violet Viola lactea, chamomile Chamaemelum nobile and golden-hair lichen Teleoschistes flavicans may be particularly vulnerable.  Trampling leads to loss of vegetation cover (e.g. heather); older stands of heather are particularly vulnerable and wet weather increases vulnerability (i.e. winter).  Trampling can be beneficial in maintaining structural diversity and a subset of adapted plants but this depends on intensity and the time of year. For a suite of plants associated with bare and poached ground (e.g. marsh clubmoss, three-lobed water crowfoot, slender centuary, and pale dog-violet) trampling in the winter may create or help maintain suitable habitat, but heavy trampling in the spring / summer may be detrimental.  Lichen flora, where present (e.g. Cladonia spp. on mature dry heath) are vulnerable to trampling. Damage can increase in the summer months when the lichen is drier and therefore more brittle. On cliff tops in areas of coastal heath, golden-hair lichen may be particularly vulnerable.  Heavy trampling (e.g. by horses or bikes) is an issue for invertebrates, especially those associated with bare ground habitats (e.g. the cuckoo bee Nomada ferruginata, shrill carder bee Bombus sylvarum, and a ruby tailed wasp Chrysis fulgida). This is an issue all year round.  Trampling can spread Phytophthora ssp. which can cause death in heather and bilberry  Fire can have catastrophic impacts, sometimes over a wide area. Vegetation recovery can be slow and a shift in vegetation community can occur. Fires can result in direct mortality to mammals, birds, herptiles and invertebrates and, where very regular or occurring over large parts of sites can result in local extinctions. Heathland wildfires tend to be concentrated between April – August which is also the period in which they are likely to do the most damage. Fires have the least impact on well managed sites (e.g. with network of fire breaks) or those with few or none of the rare species. Some species such as golden-hair lichen and cilate-strap lichen associated with cliff tops on coastal heath may be particularly vulnerable.  Recreational disturbance to ground nesting birds is well documented, and a particular issue for Annex I species such as nightjar (nests May – August), woodlark (rare in Wales, nests March – July) and Dartford warbler (nests April-July, but territorial all year).  Adders may be affected by disturbance and direct mortality. Adders may be most vulnerable when they emerge (late winter), as they can be particularly sluggish in cool weather  Winter raptor roosts may be vulnerable, especially to dogs off-leads or to birdwatchers/photographers. Raptors will tend to choose areas with deep vegetation such as deep heather or areas of dense purple moor-grass.

Fen, marsh and swamp  Whilst fen, marsh and swamp habitats are susceptible to trampling, their impenetrability can restrict access. Damage therefore tends to be intense at localised spots and along linear routes.  Fires occurring within any large build up of sedge and reed vegetation can be devastating for lower plants and fauna, particularly in the summer months when birds may be nesting. Fires outside the breeding season can have beneficial consequences such as scrub control and preventing tree encroachment.

Bogs  Whilst bog habitats are highly susceptible to trampling, their impenetrability can restrict access. Damage therefore tends to be intense at localised spots and along linear routes.  The creation of linear walking routes over bogs can eventually result in changes to hydrology, with trampled routes becoming channels along which water will flow. This reduces the volume of water retained in the bog, and can also increase localised erosion.  Erosion is most likely to occur in winter when vegetation cover is reduced.  Sphagnum mosses are the principal ‘building block’ of bog habitats, retaining large amounts of water whilst alive, and being the main component of vegetation build up in the peat layers. Sphagnum mosses are delicate in structure, and are quickly destroyed by continued abrasion. In the absence of these mosses, micro conditions within the bog are altered, and other bog species are unable to colonise.  Light levels of trampling creating small pockets of bare ground will benefit some bog plants such as species of sundew.  Areas of bog habitat that have temporarily or permanently dried out will become susceptible to fire, with the layers of peat being easily burnt, and having devastating consequences for bog fauna. This is essentially an issue during the summer months when temperatures are at their highest, and also during school holidays when more fires are started deliberately.  Once peat has been damaged by fire, it becomes vulnerable to erosion, with water forming gullies and freeze thaw action further breaking up the exposed lower layers of peat. Such areas are likely to require intervention in order to enable vegetation to re-establish.  Where vegetation is able to re-establish following fire, there may be a dominance of grassland species.  Human disturbance to breeding birds in bog habitats can result in a reduction breeding success and reduced densities  Where sites are used by wintering waterfowl or roosting raptors, disturbance may have the potential to be an issue in the winter.

Standing open water and canals  Transfer of invasive non-native plant material between sites can increase in summer months when people and dogs will venture closer to and within water.  Open water habitats tend to suffer from localised trampling damage, with access to undertake fishing, observe the water or enter the water occurring in the same spots over time.  A number of aquatic plants are susceptible to disturbance such as boating activity, e.g. floating water-plantain.  Aquatic plant communities of still water can be disturbed in the summer months when people and dogs are more likely to enter the water body.  Discarded line and nets can entangle waterfowl, and other animals, and discarded fishing weights can be swallowed and result in bird mortality.  Disturbance to aquatic mammals is little documented. The otter is well known for its secluded nature and if at all affected by high levels of disturbance, these are likely to be most damaging when the female has vulnerable young. Domestic dogs off leads around river banks will disturb breeding females.

Rivers and streams  Discarded line and nets can entangle waterfowl, and other animals, and discarded fishing weights can be swallowed and result in bird mortality.  Disturbance to aquatic mammals is little documented. The otter is well known for its secluded nature and if at all affected by high levels of disturbance, these are likely to be most damaging when the female has vulnerable young. Domestic dogs off leads around river banks will disturb breeding females.

Montane and upland habitats  Juniper Juniperus communis may be affected by Phytophthora (potentially spread through human footwear)  Trampling results in loss of vegetation cover and then to erosion. Studies have shown that trampling damage increases with slope angle, making montane habitats particularly susceptible to damage. Additional water during winter months will increase the level of damage created by each walker, whilst during the summer months the numbers of walkers enjoying montane areas will increase. A range of studies have demonstrated a widening of paths as walker volume increases.  Montane habitats on very shallow peaty or sandy substrates will quickly be lost in comparison to those on deeper soils. Difference may be greater in the summer months when shallow substrates will be drier and disintegrate more readily.  Frost damage can contribute to the degradation of already damaged substrates.  Formal or informal use of montane habitats for skiing and sledging in winter months can cause substrate abrasion and vegetation loss.  Few studies are available to identify the impact of mountain biking, but there are likely to be effects on plant communities as a result of rutting and abrasion. Diversions away from tracks are a common feature of mountain biking activity, which can spread impacts around a wider area. Rutting damage will be greater during the wetter seasons.  In the longer term, climate change may increase soil erosion with an increased occurrence of storms. Such impacts will be intensified by trampling in steeply sloped habitats. Soil erosion in one area can mean soil debris in another which can bury vegetation and damage plant communities.  The cooler climate found in montane habitats can slow the recovery rate of plants exposed to abrasion. Some montane species take tens of decades to recover from trampling damage. Lichens are particularly vulnerable.  Species at particular risk from burning include Juniper  Pine martens survive in small populations in the Welsh uplands, and one study shows a physiological stress response to human disturbance, particularly during the breeding season.

Inland rock  In some popular areas human waste can be a problem with nutrifying effects on vegetation around cliff and rock base areas  Recreational climbing can cause significant reductions in plant cover, both at the cliff base and on rock faces, due to ‘gardening’ and damage from climbers activities. Specialised rock plants and lichens are at particular risk

 Mountain climbing can be correlated with increased plant introductions, most probably through seeds being transported by climbers. Some studies have shown that bryophyte diversity is reduced in climbing spots, leaving those species most tolerant of this type of disturbance to thrive.  Seasonal effects are not documented, but it is assumed that effects will be less in winter due to a reduced level of climbing activity and the life cycles of plants within inland rock habitats.  Cliff climbing can cause disturbance to breeding birds, particularly cliff nesting raptors

Built-up areas and gardens  Disturbance at bat breeding or overwintering sites (e.g. within buildings) an issue.  Allotment gardens are an important refuge for herptiles, with disturbance during the spring and summer affecting breeding, and destruction of overwintering refuge sites resulting in direct mortality of hibernating animals.

Saline lagoons  Saline lagoons can become polluted and eutrophic from being used as unofficial rubbish dumps  Water turbidity, lagoon bed trampling and erosion of lagoon margins can all occur where there is a focus of recreational activity. Saline lagoons are hotspots for sailing, windsurfing and shellfish collection and therefore particularly vulnerable in summer months when recreational activity will be at its peak.

Coastal vegetated shingle  Vegetated shingle is a particularly sensitive and vulnerable habitat type.  Vegetated shingle is sensitive to any form of nutrient enrichment.  Very light levels of trampling, even just one trampling event, can have significant effects.  Ground nesting birds can include wader species such as ringed plover and oystercatcher. Little terns may also occur. All are vulnerable to disturbance.

Rocky shores  Algal cover can be stripped from rocks through abrasion. Algae is less able to recover out of the growing season, with the greatest recovery occurring during the spring and summer.  Trampling can result in the direct mortality of a wide range of rocky shore inhabitants, including barnacles, limpets, anemones, bivalves, worms and sponges, leading to community changes and an increase in bare rock.  Food harvesting can damage plant and animal communities

Maritime cliffs and slopes  Soft and eroding maritime cliffs will easily crumble with continued access. Scramble routes from cliff tops to the beach below can quickly develop into gullies with significant vegetation and substrate loss, which in turn encourages greater use of the same route.  Lichens associated with maritime cliffs are susceptible to damage, particularly in drier seasons when more brittle.  Cliffs are inhabited by a range of bird species with specific habitat requirements for nesting, and are vulnerable to disturbance events occurring on normally secluded sites. Most cliff

nesters are colonial birds, and large numbers can therefore be disturbed as a consequence of just one disturbance event.  Adders can be found on maritime slopes, where human disturbance or habitat damage as the adder emerges to breed can be detrimental.

Saltmarsh  The generally impenetrable nature of salt marshes provides some degree of protection for this habitat, but once an access route has been developed, it is likely to support different species to the surrounding marsh.  Pioneer saltmarsh tends to be most vulnerable to the effects of trampling, resulting in the creation of bare mud.  Discarded line and nets can entangle wading birds, and discarded fishing weights can be swallowed and result in bird mortality. This may be a particular issue outside of the summer season, when food supplies are lower and birds more likely to eat discarded rubbish.  Roosting and overwintering waders and waterfowl can gather in large numbers, and whole flocks can be disturbed by human or dog presence in close proximity. Continued disturbance causing flight will result in increased energy use over winter and avoidance of sites.

Mudflats and sandy beaches  Discarded line and nets can entangle wading birds, and discarded fishing weights can be swallowed and result in bird mortality.  Mudflats are an essential source of invertebrate food for coastal birds, and access to mudflats can result in significant changes to the invertebrate communities upon which birds feed. Direct mortality, or indirect effects arising from changes to algal growth, nutrients and mudflat topography can reduce the diversity and volume of invertebrate fauna within mudflats.  Loss of organic matter from sandy beaches, often associated with ‘beach clean ups’ can reduce the numbers of invertebrates inhabiting sandy beaches.  Roosting and overwintering waders and waterfowl can gather in large numbers, and whole flocks can be disturbed by human or dog presence in close proximity. Continued disturbance causing flight will result in increased energy use over winter, thus reducing fat reserves.

Coastal sand dunes  Foredunes are likely to suffer greater damage from trampling than the more stabilised grassy and heath dunes. A relatively small amount of trampling in mobile dunes can have a significant and disproportional effect on the habitat. Recovery is likely to be faster in the growing season.  Light levels of trampling on more stabilised dunes can create areas of bare ground, which may increase biodiversity.  Eutrophication arising from human and dog faeces can contribute to changes in species diversity and the colonisation by species not typical of unmodified dunes

 Heavier levels of trampling may lead to destruction of invertebrate burrows. Trampling can also result in direct mortality for invertebrates, of concern for particularly rare species such as the belted beauty.  Fires during the dry summer months over grassy sand dunes can lead to seedbank loss and therefore affect the long term stabilisation of the dune. Fires at this time will also affect fauna breeding within the dunes.  Human disturbance on sand dunes can both directly and indirectly affect reptile populations. Sand dune sites support both natterjack and sand lizards (Both species with very limited distributions within Wales).  Installation of boardwalks across dune to beaches can effect natural patterns of sand accretion

Mapping Seasonal Vulnerability

Overview In this section we consider how vulnerability to recreation impacts (as set out in the review section) could be captured in a map or series of maps covering the whole of Wales. Using GIS it is possible to overlay data relating to topography, pedology, habitats and species. By scoring different elements within these datasets according to their vulnerability to recreation impacts, it should be possible to derive a map that highlights geographical areas particularly vulnerable to recreation impacts. Such maps could be used to show seasonal variation (i.e. one map for each of the four seasons) or individual types of impacts (e.g. enrichment, damage, fire and disturbance).

Vulnerability maps could be a useful tool to provide a strategic overview of the whole of Wales by highlighting places where increased recreational access may be of concern, and more robust places where impacts are likely to be limited or absent. In addition to providing a guide to where issues may occur, the final maps would also be useful in targeting resources. The end product could be useful to those involved in spatial planning, access provision, access management or general policy work. Once generated the maps could be used within the GIS to compare between regions, between designated sites or plotted alongside social data to identify locations where access levels may increase and be damaging.

In order to be effective the end product must be simple to interpret and collate complex information into a single value (or a few values) for each location. The outputs must make ecological sense (i.e. highlight areas that are or could be sensitive), yet not be so detailed that the product falls down once viewed at a scale covering the whole of Wales.

Sensitivity mapping is a technique widely used by the oil industry and is often used in coastal areas - for example sensitivity mapping has been undertaken for sub-tidal habitats in Wales (e.g. Tyler- Walters and Arnold, 2008a). Another example of the approach includes maps produced by the RSPB in Scotland to show sensitivities for key bird species to wind farms (Bright et al., 2008). The use of sensitivity maps to show recreation impacts, especially at the scale addressed within this report, does however appear to be relatively novel, and there is clearly a challenge to condense the complex issues to single values that are meaningful and useful. In the rest of this report we discuss how this might be achieved. In Figure 2 we show an overview of how we envisage a sensitivity map might be derived, and each element within the figure is discussed in more detail below. Within our discussions we use a single habitat (dwarf shrub heath) to highlight issues and potential problems, to produce trial maps and show how a composite score might be derived. Some trial maps showing how the scores might be developed are given in Appendix III.

Figure 2: Overview of potential method for developing a sensitivity map for recreational impacts in Wales

Use of a grid to summarise data A grid provides a standard unit (i.e. cells or pixels) for which individual sensitivity scores can be calculated. The advantage of a grid (rather than, for example land parcels, or polygons showing habitat types) is that the basic unit from which the map is constructed is of a consistent and standard size. The choice of grid cell size is important. Cells must be small enough to provide detail and show variation, but must be large enough to be computationally viable. Potential options are 1km or 500m grid cells. The former would generate just over 22,000 cells and the latter c. 87,000 cells.

Using lowland heathland as an example to compare cell size, Figure 3 shows the frequency distribution of the percentage of each cell, by area, that is heathland for all those cells within Wales which have any lowland heathland within them. It shows that for both cell sizes, heathland covers less than 10% of the area of most cells. As the percentage of heathland within the cells increases, for both cells sizes, the percentage of cells which fall within the bands drops. However at the upper areas, greater than 60 %, the percentage of cells is significantly greater for 500m cells than for the 1 km cells, indicating that as the cell size is reduced there are a greater percentage of cells which have a large percentage of their area covered by heathland. As a consequence the use of a 500m grid would result in fewer cells having multiple habitat types within them, potentially facilitating the scoring. We therefore recommend the use of a 500m grid.

Figure 3: Comparison of the ability of 500m and 1km grid cells to capture lowland heathland. Only cells containing lowland (i.e. below 300m) heathland are included (heathland extent being extracted from Phase I data for Wales). Graph shows the frequency distribution for a series of categories representing the percentage of the cell that is heathland. Note the log scale on the y axis.

Soil Figure 2 identifies the need to score both the soil type(s) and the habitat type(s) present within a cell. The habitat score will essentially capture the vulnerability of the habitat (i.e. the plant

97 communities) to trampling and a second score will capture the vulnerability to fire. The soil type will partly determine the habitats present, and will also influence vulnerability to erosion once trampling has removed vegetation cover. It is therefore difficult to isolate soil type from habitat and the simplest approach to comparative scoring would be to derive a single score for habitats that captures erosion. The problem with this approach is that certain soil types will be particularly vulnerable to erosion once vegetation cover has disappeared. Exposed peat in degraded blanket mires is, for example, particularly vulnerable to damage. Skeletal soils are also particularly vulnerable to erosion once vegetation cover has disappeared. We therefore suggest that initially scores are derived for soil types as well as habitats, and a score is derived for enrichment and a separate score for erosion. It may be that a small number of soil types could be identified as sensitive and all others discounted. The presence of a one of the sensitive soil types within the cell would result in that cell being given a weighted score related to the area of the soil type within the cell.

The identification of potentially vulnerable soil types will need to be done through consultation with an appropriate group of soil scientists. Soil types that are highlighted can then be extracted from the National Soil Map which provides standard data for the whole of Wales, as shown in Map 1. The national soil map (or ‘Natmap’) is based on published soil maps which cover a quarter of the land at scales of 1:25 000, 1:63 360 or 1:100 000 and on reconnaissance mapping of previously unsurveyed areas. The legend shows geographic soil associations identified by the most frequently occurring soil series and by combinations of ancillary series. Sixty-seven soil groups are recognised and plotted. NatMap data are shown in Map 1 (in Appendix III).

In order to achieve a simple end map, the coverage of cells that are given a soil weighting should be checked against the habitat data to assess whether the soil dataset does provide a meaningful and useful addition.

Topography Slope angle is an important factor in determining the impact of erosion. As described in section 0 of the review, active erosion is more likely to occur on slopes with angles of 20 or above. We therefore suggest each cell with slope angles above this threshold should be scored as sensitive. Terrain data has been summarised by CCW staff for the Phase I habitat map, with data extracted for each polygon within the Phase I, summarising minimum altitude, maximum altitude and mean slope angle. These values have been derived from the OS profile dataset at 10m resolution. Using this extracted data will potentially create problems for the sensitivity map as the Phase I polygons vary markedly in size (see below), and a single slope value for the large polygons will mean that much variability in terrain will be lost. In order to capture the variability in slope angles at a standard resolution within the 500m grid it will be necessary to therefore extract slope data from CCW’s terrain model for each grid cell.

Habitat The habitat scores would be two general scores for each habitat, reflecting the sensitivity of each habitat to a) trampling and general wear and b) contamination. The scores would be determined by

98 expert opinion, with a small group of experts asked either to score each habitat (e.g. with a score between 1-5, see Maps 2 and 3) or alternatively to rank each habitat, and to do this four times to cover the four seasons. The mean score or mean rank could then be used to determine a seasonal score for each habitat. As many cells will contain multiple habitats, then the area of each habitat type will need to be calculated for each cell and the scores weighted to the area of each habitat type.

The best habitat data to use in the sensitivity mapping are the Phase I habitat data which covers the whole of Wales. Discussion with CCW staff indicates that this will be more suitable than the land cover data, which would be the alternative. The Phase I includes a long list of habitats with just over 100 different habitat types included. This level of detail is too complex for the purposes of deriving a score for each 500m cell and is also potentially too long to ask people to score. We have therefore simplified the Phase I list to derive a combined list that largely matches the habitats used in the literature review. This list is summarised in Appendix II, which gives the individual Phase I habitats within each of our broader habitat categories.

Two habitats (‘Saline lagoons’ and ‘Boundary and linear features’) are covered within the literature review and are not assigned in the table. Saline lagoons are vulnerable to physical disturbance and changes in hydrology and morphology. They also tend to act as sinks for litter and pollutants, being enclosed features. Saline lagoons are found in nine designated sites within Wales and, as a BAP priority habitat and a priority habitat on Annex 1 of the EC Habitats Directive they have been subject to dedicated survey work and mapping (Barnes, 1989, JNCC, 2007). GIS data showing individual lagoons is available for the whole of Wales and this data will allow saline lagoons to be included as a separate feature. The largest lagoon in Wales is 12 hectares, but most are less than 1 ha.

Boundary and linear features will occur in many cells and given the difficulties in recording area etc. we suggest that they are omitted from the list of habitats to include in the sensitivity mapping

Some of the habitat categories are too broad too reflect variations in sensitivity. In particular, heathland will need to be split into upland and lowland components, with lowland heathland including coastal heathland habitats. In contrast, there may be too many grassland habitats. At least for upland areas, all grassland could potentially be assigned to either improved or unimproved grassland. At the moment we suggest keeping the lowland grassland habitats separate as there are slightly different impacts and issues for each. Once the scores have been derived it may be possible to simplify the list further, merging similar habitats with similar scores as necessary. It may be that the species data can be used to capture subtle differences between habitats which have been grouped together.

In Table 11 we summarise the list of habitats for scoring by expert opinion.

Table 11: Summary list of habitats for scoring Habitat Notes Broadleaved, mixed and yew woodland Coniferous woodland Arable and horticulture Improved grassland Neutral grassland Calcareous grassland Potentially merge these in the uplands (and lowlands?) Acid grassland Bracken Lowland heath Dwarf shrub heath in the Phase I will need to be split into lowland and upland Fen, marsh and swamp Bogs Standing open water and canals Rivers and streams Inland rock Built-up areas and gardens Industrial habitats Additional category to include quarries, spoil heaps etc. within the Phase I Upland heath Coastal vegetated shingle Saltmarsh Maritime cliffs and slopes Rocky shores Coastal sand dunes Mudflats and sandy beaches Saline Lagoons Not included in Phase I

A further issue with the Phase I is that some of the polygons, particularly in the uplands, are very large: the Phase I data includes 485,000 polygons, the largest of which is over 7500ha. If a 500m grid is used, then each grid cell will be 25ha. For many habitats the polygons, unsurprisingly, typically cover much larger areas than this, as Table 12 illustrates. This isn’t in itself a particular problem for deriving the sensitivity map (it will just mean in some places large areas will have identical habitat scores). However, many polygons have secondary habitats within them. This is the case for 5,473 (i.e. just over 1%) of the polygons within the Phase I GIS, and relates to 36 different habitats (Table 13), and is particularly an issue with upland habitats. A potential solution is therefore to use a different habitat data set for the uplands. CCW also hold, for the uplands, Phase I data as points on a 1km grid. Use of this point data would capture the variation within the large polygons of the Phase I. Individual 500m cells would therefore need to be assigned habitat types relating to the nearest point or alternatively a 1km grid be used for the uplands rather than the 500m grid.

Conversely there are many situations where there is more than one habitat per grid cell or the cell appears on the edge of the phase 1 habitat map, i.e. along the coast. In these situations it is advisable to weight the habitat scores for each habitat according to the proportion of the cell covered by that habitat, rather than summing or averaging.

100

Table 12: Phase I habitat types with the largest mean areas per polygon. Habitats are ranked by mean size and only those habitats where the mean polygon size is above 5ha are shown. Mean area of Max of Count (number of Code Description polygons polygons polygons) E.1.6. raised bog 38.35 416.7 26 2 marshy grassland Molinia B.5.2 37.39 2050.18 737 dominated H.1.1 intertidal mud/sand 22.37 2502.97 1480 B.4 improved grassland 20.12 2063.94 49812 I.2.3 mine 18.27 645.64 143 E.1.7 wet modified bog 17.89 4066.66 854 E.1.8 dry modified bog 11.36 692.66 544 A.1.2. planted coniferous woodland 10.51 4756.48 15940 2 E.1.6. blanket bog 10.47 1114.87 1951 1 B.1.1 unimproved acid grassland 9.75 7588.23 13654 basic dry heath/calcareous grassland D.7 9.51 106.49 17 mosaic D.3 lichen/bryophyte heath 9.4 41.98 19 I.1.1 inland cliff 8.85 75.47 11 H.6.8 open dune 8.73 463.29 331 D.5 dry heath/acid grassland mosaic 7.97 581.27 3301 D.1.1 dry acid heath 7.47 3861.17 8743 H.2.6 salt marsh 7.24 732.1 829 E.3.3. modified flood plain mire 6.64 6.64 1 1 D.6 wet heath/acid grassland mosaic 6.47 232.73 670 J.1.1 arable 6.39 552.03 10134 H.6.5 dune grassland 5.66 257.52 452

101

Table 13: Phase I habitat types with additional habitats listed within the main habitat type Code Description Maximum number of additional habitats Total number of polygons A.1.1.1 semi-natural broadleaved woodland 1 67644 A.1.2.2 planted coniferous woodland 1 15940 A.2.1 dense scrub 4 24076 B.1 acid grassland 1 2 B.1.1 unimproved acid grassland 12 13654 B.1.2 semi-improved acid grassland 3 8939 B.3.1 unimproved calcareous grassland 8 713 B.4 improved grassland 4 49812 B.5 marshy grassland 2 24707 B.5.1 marshy grassland Juncus dominated 4 124 B.5.2 marshy grassland Molinia dominated 10 737 C.1.1 bracken 6 35301 C.2 upland species rich ledges 2 53 C.3.2 non-ruderal herb and fern 1 28 D.1.1 dry acid heath 12 8743 D.2 wet heath 4 2418 D.3 lichen/bryophyte heath 3 19 D.5 dry heath/acid grassland mosaic 6 3301 D.6 wet heath/acid grassland mosaic 3 670 E.1.6.1 blanket bog 6 1951 E.1.7 wet modified bog 9 854 E.1.8 dry modified bog 7 544 E.2 flush and spring 1 17 E.2.1 acid/neutral flush 7 8594 E.2.2 basic flush 3 154 E.3.1 valley mire 1 286 E.3.2 basin mire 3 372 E.4 bare peat 3 68 F.1 swamp 1 2222 G.1 standing water 1 9598 I.1 natural rock exposure 2 48 I.1.1 inland cliff 2 11 I.1.1.2 basic inland cliff 1 23 I.1.2 scree 3 212 I.1.2.1 acid/neutral scree 5 795 I.1.4 other rock exposure 3 149 I.2.1 quarry 2 1094 I.2.2 spoil 3 1447 NA #N/A 2 14720

102

Fire A fire severity index is derived by the Met Office on a 10km grid basis for the whole of Wales on a daily basis12. This index is based on a system developed in Canada and takes into account moisture conditions at various levels through the soil; weather elements (temperature, relative humidity, wind speed and rainfall); the long term effects of the weather on ground conditions; and the severity of potential fire, and identifies fire severity according to a range of weather scenarios such as hot dry summers and short spells of hot windy weather. Data for each 10km cell are available on a daily basis for previous years and therefore it will be possible to derive a mean value for each 10km cell, for each season, using data for a given time period, e.g. over a five year time period.

The fire severity index essentially captures the risk of a fire taking place, and therefore does not capture the impact that a fire would have on a given area. The impact of fire on different plant communities would therefore also need to be captured, through expert opinion, on a habitat by habitat basis. Such a score could be generated using the same group of experts as the general damage score for habitats described above. This score, describing the impact of fire, could then be combined with the severity index by multiplying the two values together to give a single value that captures both the risk and severity of fire.

Plant Species Individual plant species may be vulnerable to trampling, harvesting or nutrient enrichment or fire. Species records would need to be gathered for a list of species agreed through expert opinion, with the list representing key species or particularly vulnerable ones. An exhaustive list is clearly too complicated. Selecting a suite of representative and key species, including a range of higher and lower plant taxa, niches and habitats, will be challenging. We suggest that a provisional list of species is produced and circulated to a range of botanists and ecologists with a good understanding of the welsh flora. The list will need to identify impacts for each species (i.e. damage, contamination or fire). Once a finalised list has been derived, it will then be necessary to determine a score for each species for each impact. It may be possible to give each species a standard score, or it may be that certain species are given a greater weighting, for example by asking each consultee to rank the species within the list (allowing the final score for each species to be derived from a its mean rank).

Table 14 shows a provisional list of potentially sensitive plant species associated with lowland heathland. The list contains ten species, most of which have very limited distributions. Three species (spotted rock-rose and the two lichen species) can occur in the same locations (e.g. the Llyn peninsula) and are coastal in their distribution. A number of the other species (such as the water- crowfoot and slender centuary) are associated with poached ground and often occur in vehicle ruts and tracks where the ground is churned up over the winter. Using this list would mean that grid cells in particular areas such as the Llyn peninsula and some other coastal sites (where a number of the species occur together) would come out with particularly high sensitivity scores. Within extensive areas of heathland, cells that have particular micro-habitats (such as tracks with poached ground) will also have high sensitivity scores.

12 See: http://www.ccw.gov.uk/enjoying-the-country/countryside-access-map/the-fire-severity-index.aspx and http://www.ccw.gov.uk/pdf/Fire_Severity_Index_Leaflet.pdf

103

Species data are held for Wales by COFNOD and there are also data available from NBN. For each species it will be necessary to define the range of species data used and how it is included within the GIS. For example, species records may cover a wide time period and will vary in the accuracy with which locations are recorded. For most species, it would be best to only use recent records and exclude data where the grid reference or location is not sufficiently specific (e.g. 4 figure grid references).

104

Table 14: Provisional list of heathland plant species to include within the sensitivity map. This table was produced following discussion with Jan Sherry (CCW). Common name Latin Impact of recreation

Three-lobed water-crowfoot Ranunculus tripartitus trampling in spring / summer

Slender centuary Cicendia filiformis trampling in spring / summer

Spotted rock-rose Tuburaria gutatta burning

Ciliate strap-lichen Heterodermia leucomela burning and trampling

Pale Dog-violet Viola lactea trampling and enrichment (e.g. from dog fouling)

Chamomile Chamaemelum nobile enrichment

Pillwort Pilularia globulifera enrichment

Golden hair lichen Teloschistes flavicans trampling, fire and enrichment

Dwarf rush Juncus capitatus trampling (in winter & spring), enrichment

Animal Species Data on particular animal species will be similar to plants. It will be necessary to derive an indicative list of birds, mammals and invertebrates and for each species identify whether the impact is from disturbance, fire or both. For birds and mammals the list will be relatively straight forward and can be extracted from the literature review (for example see page 77 for birds), potentially circulating the list to species experts within CCW for confirmation. As with plants it may be necessary to give different species different scores. It may also be simpler to group species, such as cliff-nesting seabirds, tern colonies, wintering geese and swans. In such cases it would then be straightforward to identify sites rather than use point data relating to species records.

The literature review and subsequent discussion with CCW ecologists has identified the following provisional species associated with heathland:

 nightjar  marsh fritillary  woodlark  silver-studded blue  Dartford warbler  a cuckoo bee Nomada ferruginata  chough  shrill carder bee  adder  a ruby tailed wasp Chrysis fulgida

Most of the species are very rare or have limited distributions that also encompass other habitats. For example, chough occur in coastal areas and in Snowdonia, nest in caves and feed in grassland and sand dunes and are therefore loosely associated with heathland. Within Wales, nightjar are

105 largely limited to clearfell areas within conifer plantations. The species data for animals will therefore not necessarily be closely linked to habitat and will ensure a varied score across areas of uniform habitat.

For each species it will be necessary to carefully assess what data to use to identify which cells within the grid hold or potentially hold the species. As with the plants, species datasets will cover a range of years and the data include truncated grid references. For species such as nightjar, national survey data (for nightjar the last national survey was in 2004) will provide standard coverage across all sites. For some species it may be necessary to buffer individual records as adjoining grid cells may have the potential to support the species or be within range.

Overall Scores Based on the above it would be possible to derive an individual overall score for each cell in each season (see Map 4). This overall score would be the sum of scores for contamination, damage, fire and disturbance. These different component scores could also be plotted individually to produce a separate map for each season for contamination, damage, fire and disturbance. Once the species lists and other scores have been finalised, it may be necessary to weight the scores for each type of impact to ensure that the overall score is not overly influenced by one type of impact. For example, if the scores for contamination ranged from 0-2 and the range of scores for disturbance ranged from 0-50 due to the number of species included, it might be necessary to weight the impact scores to ensure a balanced overall score. We summarise how the individual scores are calculated below:

Contamination The score for each grid cell would be the sum of:

 The score for soil type and enrichment (possibly not necessary),  The general score for each habitat and impacts of contamination  The presence of different key plant species vulnerable to contamination (each species potentially with an individual score)

Damage The score for each grid cell would be the sum of:

 The score for soil type and erosion  The general score for each habitat and impacts of damage  The presence of different key plant species vulnerable to damage such as trampling (each species potentially with an individual score)

Fire Score for each grid cell would be the product of:

 mean value of the fire severity index over a fire year period  general score for each habitat on the impact of fire

106

Disturbance The score for each grid cell would be the sum of:

 The presence of different key animal species vulnerable to disturbance (each species potentially with an individual score)

107

References Addessi, L. (1994) Human disturbance and long-term changes on a rocky intertidal community. Ecological Applications, 4, 786-797. Ahlgren, I. (1974) The affects of fire on soil organisms. Fire and Ecosystems (ed T. T. K. C. E. Ahlgren). Academic press, New York. Aked, T. (1984) Soil invertebrate fauna of moorland habitats and their relationship to fire damaged areas. North York Moors National Pak, Helmsley. Alchin, E. (1997) Vegetation dynamics following management burning of lowland heath. University of Liverpool, Liverpool. Alessa, L., Bennett, S. M. & Kliskey, A. D. (2003) Effects of knowledge, personal attribution and perception of ecosystem health on depreciative behaviors in the intertidal zone of Pacific Rim National Park and Reserve. Journal of Environmental Management, 68, 207-218. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005a) Habitat Section 7 Coastal Saltmarsh. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005b) Habitat Section 8 Coastal Sand Dunes. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005c) Habitat Section 9 Vegetated Shingle. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005d) Habitat Section 15 Lowland calcareous grassland. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005e) Habitat Section 17 Lowland heathland. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005f) Habitat Section 20 Lowland wood pasture and parkland. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005g) Habitat Section 30 Upland mixed ashwoods. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005h)

108

Habitat Section 31 Upland oakwood. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Alexander, K., Archer, M., Colenutt, S., Denton, J., Falk, S., Godfrey, A., Hammond, P., Ismay, J., Lee, P., Macadam, C., Morris, M., Murray, C., Plant, C., Ramsay, A., Schulten, B., Shardlow, M., Stewart, A., Stubbs, A., Sutton, P., Telfer, M., Wallace, I., Willing, M. & Wright, R. (2005i) Habitat Section 32 Wet woodland. Managing Priority Habitats for Invertebrates. Buglife - The Invertebrate Conservation Trust. Allchin, E. A. (1997) Vegetation dynamics following management burning of lowland heathland. University of Liverpool. Allen, S. (1964) Chemical aspects of heather burning. Journal of Applied Ecology, 1, 247-367. Andersen, A. (1991) Responses of ground-foraging ant communities to three experimental fire regimes in a savanna forest of tropical Australia. Biotropica, 23, 575-585. Andersen, U. V. (1995) Resistance of Danish coastal vegetation types to human trampling. Biological Conservation, 71, 223-230. Anderson, P. (1986) Accidental moorland fires in the Peak District: A study of their incidence and ecological implications. Peak Park Joint Planning Board, Bakewell. Anderson, P. (1997) Fire damage on blanket mires. Blanket mire degradation: Causes, consequences and challenges (eds J. H. Tallis, R. Meade & P. D. Hulme). British Ecological Society. Mires Research Group, Manchester. Anderson, P., Tallis, J. & Yalden, D. (1997) Restoring moorland: Peak District Moorland Management Project Phase III report. Peak Park Joint Planning Board, Bakewell. Arndt, E., Aydin, N. & Aydin, G. (2005) Tourism impairs tiger beetle (Cicindeldae) populations – a case studyin a Mediterranean beach habitat. Journal of Insect Conservation, 9, 201-206. Arnesen, T. (1999) Vegetation dynamics following trampling in rich fen at Sølendet, Central Norway; a 15 year study of recovery. Nordic J. of Botany, 19, 313-327. Arroyo, B. & Razin, M. (2006) Effect of human activities on bearded vulture behaviour and breeding success in the French pyrenees. Biological Conservation, 128, 276-284. Asken Ltd (2008) Bowland Open Access: People and hen harrier monitoring project. Lancashire County Council. Atkinson, P. W., Buckingham, D. & Morris, A. J. (2004) What factors determine where invertebrate- feeding birds forage in dry agricultural grasslands? Ibis, 146, 99-107. Backshall, J., Manley, J. & Rebane, M. (2001) The upland management handbook. English Nature, Peterborough. Baines, D. & Richardson, M. (2007) An experimental assessment of the potential effects of human disturbance on Black Grouse Tetrao tetrix in the North Pennines, England. Ibis, 149, 56-64. Bakker, J. P. (1998) The impact of grazing on plant communities. Grazing and conservation management. (eds M. F. WallisDeVries, J. P. Bakker & S. E. Van Wieren). Kluwer Academic Publishing, Dordrescht. Bally, R. & Griffiths, C. L. (1989) Effects of human trampling on an exposed rocky shore. International Journal of Environmental Studies, 34, 115 - 125. Bamber, F., Batten, S. & Bridgewater, N. (1993) Design criteria for the creation of brackish lagoons. Biodiversity and Conservation, 2, 127-137. Banks, P. B. & Bryant, J. V. (2007) Four-legged friend of foe? Dog-walking displaces native birds from natural areas. Biology Letters, 3, 611-613. Barclay-Estrup, P. (1974) Arthropod populations in a heathland as related to cyclical chages in the vegetation. Entomologist's Monthly Magazine, 109, 79-84. Barnard, A. (2003) Getting the Facts - Dog Walking and Visitor Number Surveys at Burnham Beeches and their Implications for the Management Process. Countryside Recreation, 11, 16 - 19. Barnes, R. (1989) The coastal lagoons of Britain; An overview and conservation appraisal. Biological Conservation, 49, 295-313.

109

Baudains, T. P. & Lloyd, P. (2007) Habituation and habitat changes can moderate the impacts of human disturbance on shorebird breeding performance. Animal Conservation, 10, 400-407. Baur, A. & Baur, B. (1990) Are road barriers to dispersal in the land snail Arianta arbustorum? Canadian J. of Zoology, 68, 613-617. Bayfield, N. (1979) Recovery of four montane heath communities on Cairngorm, Scotland from disturbance by trampling. Biological Conservation, 15, 165-179. Bayfield, N. G. (1971) Some effects of walking and skiing on vegetation at Cairngorm. Scientific Management of Animal and Plant Communities for Conservation (eds E. Duffey & S. A. Watt), pp. 469-485. Blackwell Scientific Publications, Oxford. Bayfield, N. G. (1973) Use and deterioration of some Scottish hill paths. Journal of Applied Ecology, 10, 635-44. Bayfield, N. G. (1987) Approaches to reinstatement of damaged footpaths in the Three Peaks area of the Yorkshire Dales National Park. Agriculture and conservation in the hills and uplands. (ITE Symposium, 23). (eds M. Bell & R. G. H. Bunce). NERC/ITE, Grange-over-Sands. Bayfield, N. G. & Aitken, R. (1992) Managing the impacts of recreation on vegetation and soils: a review of techniques. INSTITUTE OF TERRESTRIAL ECOLOGY (Natural Environment Research Council). Bayfield, N. G. & Brookes, B. S. (1979) Effects of repeated use of an area of heather Calluna vulgaris (L.) hull moor at Kindrogan, Scotland, for teaching purposes. Biological Conservation, 16, 31- 41. Bayfield, N. G., Urquhart, U. H. & Cooper, S. M. (1981) Susceptibility of four species of Cladonia to disturbance by trampling in the Cairngorm mountains, Scotland. Journal of Applied Ecology, 18, 303-310. Beale, C. M. (2007) Managing visitor access to seabird colonies: a spatial simulation and empirical observations. Ibis, 149, 102-111. Beale, C. M. & Monaghan, P. (2004a) Behavioural responses to human disturbance: a matter of choice? Animal Behaviour, 68, 1065-1069. Beale, C. M. & Monaghan, P. (2004b) Human disturbance: people as predation-free predators? Journal of Applied Ecology, 41, 335-343. Beale, C. M. & Monaghan, P. (2005) Modeling the Effects of Limiting the Number of Visitors on Failure Rates of Seabird Nests. Conservation Biology, 19, 2015-2019. Beauchamp, K. & Gowing, M. (2003) A quantitive assessment of human trampling effects on a rocky intertidal community. Marine Environmental Research, 7, 279-293. Bechet, A., Giroux, J. F. & Gauthier, G. (2004) The effects of disturbance on behaviour, habitat use and energy of spring staging snow geese. Journal of applied ecology, 41, 689-700. Bechet, A., Giroux, J. F., Gauthier, G., Nichols, J. D. & Hines, J. E. (2003) Spring hunting changes the regional movements of migrating greater snow geese. Journal of Applied Ecology, 40, 553- 564. Beck, A. & Vogl, R. (1972) The effects of spring burning on rodent populations in a brush prairie savanna. Journal of Mammology, 53, 336-346. Bell, J., Wheater, C. & Cullen, W. (2001) The implications of grassland and heathland management for the conservation of spider communities: A review. Journal of the Zoological Society of London, 255, 377-387. Bell, K. L. & Bliss, L. C. (1973) Alpine disturbance studies, Olympic National Park, USA. Biological Conservation, 5, 25-32. Bellan, G. L. & Bellan-Santini, D. R. (2001) A review of littoral tourism, sport and leisure activities: consequences on marine flora and fauna. Aquatic Conservation-Marine and Freshwater Ecosystems, 11, 325-333. Benedict, R. A. & Billeter, M. C. (2004) Discarded bottles as a cause of mortality in small vertebrates. Southeastern Naturalist, 3, 371-377.

110

Benninger-Truax, M., Vankat, J. L. & Schaefer, R. L. (1992) Trail corridors as habitat and conduits for movement of plant species in Rocky Mountain National Park, Colorado, USA. Landscape Ecology, 6, 269-278. Berdowski, J. J. M. & Siepel, H. (1988) Vegetative regeneration of Calluna vulgaris at different ages and fertiliser levels. Biological Conservation, 46, 85-94. Best, L. B. (1979) Effects of fire on a field sparrow population. American Midland Naturalist, 101, 434-442. Blackman, G. E. & Rutter, A. J. (1950) Physiological and ecological studies in the analysis of plant environment. V. an assessment of the factors controlling the distribution of the Bluebell (Scilla non-scripta) in different communities. Ann. Bot, 14, 487-520. Blackmer, A. L., Ackerman, J. T. & Nevitt, G. A. (2004) Effects of investigator disturbance on hatching success and nest-site fidelity in a long-lived seabird, Leach’s storm-petrel. Biological Conservation, 116, 141-148. Blanco, G., Fargallo, J. A., Tella, J. L. & Cuevas, J. A. (1997) Role of buildings as nest-sites in the range expansion and conservation of choughs Pyrrhocorax pyrrhocorax in Spain. Biological Conservation, 79, 117-122. Blumstein, D. T. (2003) Flight-initiation distance in birds is dependent on intruder starting distance. Journal of Wildlife Management, 67, 852-857. Blumstein, D. T., Anthony L. L., Harcourt, R. & Ross, G. (2003 ) Testing a key assumption of wildlife buffer zones: is flight initiation distance a species-specific trait? BIOLOGICAL CONSERVATION, 110, 97-100. Blumstein, D. T., Fernandez-Juricic, E., Zollner, P. A. & Garity, S. C. (2005) Inter-specific variation in avian responses to human disturbance. Journal of Appled Ecology. Bolduc, F. & Guillemette, M. (2003) Human disturbance and nesting success of Common Eiders: interaction between visitors and gulls. BIOLOGICAL CONSERVATION, 110, 77-83. Bonner, C. & Agnew, A. D. Q. (1983) Soil phosphorous as an indicator of canine faecal pollution in urban recreation areas. Environmental Pollution (Series B), 6, 145-156. Bonte, D. (2005) Anthropogenic induced changes in nesting densities of the dune-specialised digger wasp Bembix rostrata (Hymenoptera: Sphecidae). European Journal of Entomology, 102, 809-812. Bonte, D. & Hoffman, M. (2005) Are coastal dune management actions for biodivesity restoration and conservation underpinned by internationally published scientific research? International conference on nature restoration practices in European coastal habitats (eds J.-L. Herrier, J. Mees, A. Salman, J. Syes, H. Van Nieuwenhuyse & I. Dobbelaere), pp. 165-178. VLIZ, Koksijde, Belgium. Boorman, L. A. (2003) Saltmarsh Review. An overview of coastal saltmarshes, their dynamic and sensitivity characteristics for conservation and management. JNCC. Boorman, L. A. & Fuller, R. M. (1977) Studies on the impact of paths on dune vegetation at Winterton, Norfolk, England. Biological Conservation, 12, 141-148. Borkowski, J. (2001) Flight behaviour and observability in human-disturbed sika deer. Acta Theriologica, 46, 195-206. Brawn, J. D., Robinson, S. K. & Thompson III, F. R. (2001) The Role of Disturbance in the Ecology and Conservation of Birds. Annual Review of Ecology and Systematics, 32, 251-276. Brian, M. V., Mountford, M. D., Abbott, A. & Vincent, S. (1976) The changes in ant species distribution during ten years post-fire regeneration of a heath. Journal of Animal ecology, 45, 115-133. Bright, A., Reynolds, G. R., Innes, J. & Waas, J. R. (2003) Effects of motorised boat passes on the time budgets of New Zealand dabchick, Poliocephalus rufopectus. Wildlife Research, 30, 237-244. Bright, J., Langston, R., Bullman, R., Evans, R., Gardner, S. & Pearce-Higgins, J. (2008) Map of bird sensitivities to wind farms in Scotland: A tool to aid planning and conservation. Biological Conservation, 141, 2342-2356.

111

Brooker, M. & Rowley, I. (1991) The impact of wildfire on the nesting behaviour of birds in heathland. Australian Wildlife Research, 18, 249-263. Brown, E. G. & Prior, A. (1998) Recreational disturbance to breeding seals and seabirds on Mousa, SSSI: report to Scottish Natural Heritage on contract no: HT/97/98/33, Scottish Natural Heritage, Scotland. Brown, P. & Taylor, R. (1999) Effects of trampling by humans on animals inhabiting coralline algal turf in the rocky intertidal. Journal of Experimental Marine Biology and Ecology, 235, 45-53. Buckley, R. (2004) Environmental impacts of Ecotourism, CABI, Wallingford, Oxfordshire. Buffington, J. (1966) Soil arthropod populations of the New Jersey pine barrens as affected by fire. Annals of the Entomological Society of America, 60, 530-535. Bull, S. E. (1998) The impact of dogs on National Trust properties. Estates Dept., National Trust, Cirencester. Bullock, J. M. & Webb, N. R. (1995) Responses to severe fires in heathland mosaics in Southern England. Biological Conservation, 73, 207-214. Burfield, I. J. & Brooke, M. d. L. (2005) The decline of the Ring Ouzel Turdus torquatus in Britain: evidence from bird observatory data. Ringing & Migration, 22, 199 - 204. Burger, J. (1991) Foraging behaviour and the effetcs of human disturbance on the Piping Plover (Chararius melodus). J. Coastal Res, 7, 39. Burger, J. (1998) Effects of motorboats and personal watercraft on flight behavior over a colony of Common Terns. Condor, 100, 528-534. Burger, J. (2003) Personal watercraft and boats: Coastal conflicts with common terns. Lake and Reservoir Management, 19, 26-34. Burton, N. H. & Armitage, M. J. (2005) Differences in the diurnal and nocturnal use of intertidal feeding grounds by Redshank Tringa totanus. Bird Study, 52, 120-128. Burton, N. H., Rehfisch, M. M. & Clark, N. A. (2002a) Impacts of disturbance from construction work on the densities and feeding behavior of waterbirds using the intertidal mudflats of Cardiff Bay, UK. Environ Manage, 30, 865-71. Burton, N. H. K. (2007) Landscape approaches to studying the effects of disturbance on waterbirds. Ibis, 149, 95-101. Burton, N. H. K., Armitage, M. J. S., Musgrove, A. J. & Rehfisch, M. M. (2002b) Impacts of man-made landscape features on numbers of estuarine waterbirds at low tide. Environmental Management, 30, 857-864. Byfield, A. & Pearman, D. (1996) Dorset's disappearing flora. Changes in the distribution of Dorset's rarer heathland species 1931 to 1993, Plantlife. Cadbury, C. (1992a) Grazing and other management of upand vegetation: A review with special reference to birds. RSPB, Sandy. Cadbury, C. (1992b) Grazing and other management of upland vegetation: A review with special reference to birds. RSPB, Sandy. Calado, M. (1996) Little Tern status and conservation at Ria Formosa Natural Park, Algarve, Portugal. Colonial Waterbirds., 19, 78-80. Carney, K. M. & J., S. W. (1999) A review of human disturbance effects on nesting colonial waterbirds. WATERBIRDS, 22, 68-79. Carney, K. M. & Sydeman, W. J. (1999) A review of human disturbance effects on nesting colonial waterbirds. WATERBIRDS, 22, 68-79. Casu, D., Ceccherelli, G. & Castelli, A. (2006a) Immediate Effects of Experimental Human Trampling on Mid-Upper Intertidal benthic Invertebrates at the Asinara Island MPA (NW Mediterranean). Hydrobiologia, 555, 271-279. Casu, D., Ceccherelli, G., Curini-Galletti, M. & Castelli, A. (2006b) Short-term effects of experimental trampling on polychaetes of a rocky intertidal substratum (Asinara Island MPA, NW Mediterranean). Scientia Marin, 70, 179-186.

112

Catry, T., Ramos, J. A., Catry, I., Allen-revez, M. & Grade, N. (2004) Are salinas a suitable alternative breeding habitat for Little Terns Sterna albifrons? Ibis, 146, 247-257. CCW & FC (2009) Welsh Outdoor Recreation Survey 2008: Headline Results. Countryside Council for Wales; Forestry Commission. Cessford, G. R. (1995) Off-Road Impacts of Mountain Bikes. A review and discussion. Science and Research series no. 29. Department of Conservation, Wellington, New Zealand. Chandrasekara, W. U. & Frid, C. L. J. (1996) Effects of human trampling on tidalflat infauna. Aquatic Conservation: Marine and Freshwater Ecosystems, 6, 299-311. Chanin, P. (2003) Ecology of the European Otter. Conserving Natura 2000 Rivers, Ecology Series. English Nature, Peterborough. Chapman, S. B., Rose, R. J. & Basanta, M. (1989) Phosphorus adsorption by soils from heathlands in southern England in relation to successional change. Journal of Applied Ecology, 26, 673-680. Chappell, H. G., Ainsworth, J. F., Cameron, R. A. D. & Redfern, M. (1971) The effect of trampling on a chalk grassland ecosystem. Journal of Applied Ecology, 8, 869-82. Christensen, S. N. & Johnsen, I. (2001) The lichen rich coastal heath vegetation on the isle of Anholt, Denmark - conservation and management. Journal of Coastal Conservation, 7, 13-22. Clarke, M. (2008) Catering for the needs of fauna in fire management:science or just wishful thinking? Wildlife research, 35, 385-394. Cole, D. N. (1978) Estimating the Susceptibility of Wildland Vegetation to Trailside Alteration. The Journal of Applied Ecology, 15, 281-286. Cole, D. N. (1986) Resource impacts caused by recreation. The President's Commission on Americans Outdoors: a literature review. U.S. Government Printing Office, Washington, DC. Cole, D. N. (1987) Effects of three seasons of experimental trampling on five montane forest communities and a grassland in western Montana, USA. Biological Conservation, 40, 219- 244. Cole, D. N. (1995) Experimental Trampling of Vegetation. I. Relationship Between Trampling Intensity and Vegetation Response. Journal of Applied Ecology, 32, 203-214. Coleman, R. (1981) Footpath erosion in the English Lake District. Applied Geography, 1, 121-131. Coleman, R. A., Salmon, N. A. & Hawkins, S. J. (2003) Sub-dispersive human disturbance of foraging oystercatchers Haematopus ostralegus. Ardea, 91, 263-268. Colman, J. E., Jacobsen, B. W. & Reimers, E. (2001) Summer response distances of Svalbard reindeer Rangifer tarandus platyrhynchus to provocation by humans on foot. Wildlife Biology, 7, 275- 283. Coombes, E. G. (2007) The effects of climate change on coastal recreation and biodiversity. School of Environmental Sciences. University of East Anglia, Norwich. Cope, D. R., Pettifor, R. A., Griffin, L. R. & Rowcliffe, M. J. (2003) Integrating farming and wildlife conservation: the Barnacle Goose Management Scheme. Biological Conservation, 110, 113- 122. Coulson, J., Fielding, C. & Goodyear, S. (1992) The management of moorland areas to enhance their nature conservation interest. Report No 134. Joint Nature Conservation Committee, Peterborough. Covey, R. & Laffoley, D. d. A. (2002) Maritime State of Nature Report for England: getting onto an even keel. English Nature, Peterborough. Cowie, N. R., Sutherland, W. J., Ditlogo, M. K. M. & James, R. (1992) The effects of conservation management of reed beds II. The flora and litter disappearance. Journal of Applied Ecology, 29, 277-284. Crick, H., Dudley, C. & Glue, D. (2003) The Nest Record Scheme Handbook, BTO. Cryer, M., Linley, N. W., Ward, R. M., Stratford, J. O. & Randerson, P. F. (1987) Disturbance of overwintering wildfowl by anglers at two reservoir sites in South Wales. Bird Study, 34, 191- 199.

113

Cunningham, P. N., Hawkins, S. J., Jones, H. D. & Burrows, M. T. (1984) The geographical distribution of Sabellaria alveolata (L.) in England, Wales and Scotland, with investigations into the community structure of and the effects of trampling on Sabellaria alveolata colonies. (ed N. C. Council). University of Manchester, Department of Zoology, Peterborough. Curr, R., Koh, A., Edwards, E., Williams, A. & Davies, P. (2000) Assessing anthropogenic impact on Mediterranean sand dunes from aerial digital photography. Journal of Coastal Conservation, 6, 15-22. Currall, J. (1981) Some effects of management by fire on wet heath vegetation in Western Scotland. Aberdeen. PhD Thesis, University of Aberdeen. Dale, D. & Weaver, T. (1974) Trampling Effects on Vegetation of the Trail Corridors of North Rocky Mountain Forests. The Journal of Applied Ecology, 11, 767-772. Davidson, N., Rothwell, P., Wader Study Group., Joint Nature Conservation Committee (Great Britain). & Royal Society for the Protection of Birds. (1993) Disturbance to waterfowl on estuaries, RSPB ; Wader Study Group, Sandy, Beds S.l. de Boer, H. Y., van Breukelen, L., Hootsmans, M. J. M. & van Wieren, S. E. (2004) Flight distance in roe deer Capreolus capreolus and fallow deer Dama dama as related to hunting and other factors. Wildlife Biology, 10, 35-41. de Gouverain, R. (1996) Indirect impacts of soil trampling on tree growth and plant succession in the North Cascade Mountains of Washington. Biological Conservation, 75. Delaney, D. K., Grubb, T. G., Beier, P., Pater, L. L. M. & Reiser, H. (1999) Effects of Helicopter Noise on Mexican Spotted Owls. The Journal of Wildlife Management, 63, 60-76. Delettre, Y. (1995) Heathland fires increase species diversity-an example in terrestrial Chironomidae (Diptera). Landscape and Urban Planning, 31, 259-268. Denis, T. (2003) Effects of human for traffic on the standing stock, reproduction and size structure of populations of the intertidal rockweed Silvetia compressa (O. Fucales). Californian State University, Fullerton, USA. Ditlhogo, M., James, R., Laurence, B. & Sutherland, W. (1992) The effects of conservation management of reed beds I. The Invertebrates. Journal of Applied Ecology, 29, 265-276. Doody, P. J. & Randall, R. (2003) Guidance for the management of coastal vegetated shingle. English Nature. Drewitt, A. (1999) Disturbance effects of aircraft on birds. Birds Network Information Note, pp. 14. English Nature, Peterborough. Driscol, D. & Henderson, M. (2008) How many common reptile species are fire specialists? A replicated natural experiment highlights the predictive weakness of a fire succession model. Biological Conservation, 141, 460-471. Duffey, E. (1975) The effects of human trampling on the fauna of grassland litter. Biological Conservation, 7, 225-274. Dunn, B. A. (1984) Recreation effects on forest soil and vegetation: Research synopsis and selected bibliography. Technical Paper. Dept. Forestry, Clemson Univ., USA. Dunn, E. K. (1975) The role of environmental factors in the growth of tern chicks. Journal of Animal Ecology, 44, 743-754. Duran, L. & Castilla, J. (1989) Variation and persistence of the middle rocky intertidal community of central Chile, with and without human harvesting. Marine Biology, 103, 555-562. Eason, P. K., Sherman, P. T., Rankin, O. & Coleman, B. (2009) Factors Affecting Flight Initiation Distance in American Robins %J Journal of Wildlife Management. 70, 1796-1800. Eaton, M. F., Brown, A. F., Noble, D. G., Musgrove, A. J., Aebischer, N. J., Gibbons, D. W., Evans, A. & Gregory, R. D. (2009) Birds of Conservation Concern 3: the population status of birds in the United Kingdom, Channel Islands and the Isle of Man. British Birds, 102, 296-341. Edgar, P. (2002) The effects of public access on amphibians and reptiles. an assessment of the potential effects of increased public access due to the introduction of the countryside and rights of way act 2000. CCW Contract Science, Bangor.

114

Ellison, L. (1975) Density of Alaskan spruce grouse before and after fire. The Journal of Wildlife Management, 39, 468-471. Enggist-Dublin, P. & Ingold, P. (2003) Modelling the impact of different forms of wildlife harassment, exemplified by a quantitative comparison of the effects of hikers and paragliders on feeding and space use of chamois Rupicapra rupicapra. Wildlife Biology, 9, 37-45. English Nature (2001) State of nature: The upland challenge. English Nature, Peterborough. English Nature (2003) Common standards monitoring guidance for sand dune habitats. Joint Nature Conservation Committee, Peterborough. Environment Agency (Undated) Funeral practices, spreading ashes and caring for the environment. www.environment-agency.gov.uk (ed E. Agency). Environment Agency, Bristol. Erdos, S., Baldi, A. & Batary, P. (2009) Nest-site selection and breeding ecology of skylarks Alauda arvensis in Hungarian farmland. Bird Study, 56, 259-263. Evans, C. & Allen, S. (1971) Nutrient losses in smoke produced during heather burning. Oikos, 22, 149-154. Fairbanks, W. S. & Tullous, R. (2002) Distribution of pronghorn (Antilocapra americana Ord) on Antelope Island State Park, Utah, USA, before and after establishment of recreational trails. Natural Areas Journal, 22, 277-282. Farrell, P. (1998) Environmental impacts of baitdigging. Environmental Research for Estuary Management: the Chichester Harbour Approach. Environmental Research for Estuary Management: the Chichester Harbour Approach. (ed A. de Potiers). Proceedings of a Seminar held on 27 April 1988. Chichester Harbour Conservancy. Farris, M. A. (1998) The effects of rock climbing on the vegetation of three Minnesota cliff systems. Canadian Journal of Botany, 76, 1981-1990. Fernandez-Juricic, E., Jimenez, M. D. & Lucas, E. (2001) Alert distance as an alternative measure of bird tolerance to human disturbance: implications for park design. Environmental Conservation, 3, 263 - 269. Fernandez-Juricic, E., Jimenez, M. D. & Lucas, E. (2002) Factors affecting intra- and inter-specific variations in the difference between alert distances and flight distances for birds in forested habitats. Canadian Journal of Zoology-Revue Canadienne De Zoologie, 80, 1212-1220. Fernandez-Juricic, E., Sallent, A., Sanz, R. & Rodriguez-Prieto, I. (2003) Testing the risk-disturbance hypothesis in a fragmented landscape: Nonlinear responses of House Sparrows to humans. Condor, 105, 316-326. Fernandez-Juricic, E. & Schroeder, N. (2003) Do variations in scanning behavior affect tolerance to human disturbance? Applied Animal Behaviour Science, 84, 219-234. Fernandez-Juricic, E., Venier, M. P., Renison, D. & Blumstein, D. T. (2005) Sensitivity of wildlife to spatial patterns of recreationist behavior: A critical assessment of minimum approaching distances and buffer areas for grassland birds. Biological Conservation, 125, 225-235. Ficetola, G. F., Sacchi, R., Scali, S., Gentilli, A., De Bernardi, F. & Galeotti, P. (2007) Vertebrates respond differently to human disturbance: implications for the use of a focal species approach. Acta Oecologica, 31, 109-118. Finney, S. K. (2004) The effect of recreational disturbance on two upland breeding birds, the golden plover Pluvialis apricaria and the dunlin Calidris alpina, Royal Society for the protection of Birds, Sandy, Beds. Finney, S. K., Pearce-Higgins, J. W. & Yalden, D. W. (2005a) The effect of recreational disturbance on an upland breeding bird, the golden plover Pluvialis apricaria. Biological Conservation, 121, 53-63. Finney, S. K., Pearce-Higgins, J. W. & Yalden, D. W. (2005b) The effect of recreational disturbance on an upland breeding bird, the golden plover Pluvialis apricaria Biological Conservation, 121, 53-63. Fisher, J. & Wilkinson, L. (205) The response of mammals to forest fire and timber harvest in the North American boreal forest. Mammal Review, 35, 51-81.

115

Fitzpatrick, S. & Bouchez, B. (1998) Effects of recreational disturbance on the foraging behaviour of waders on a rocky beach. Bird Study, 45, 157-171. Fletcher, H. & Frid, C. L. J. (1996) Impact and management of visitor pressure on rocky intertidal algal communities. Aquatic Conservation: Marine and Freshwater Ecosystems, 6, 287-297. Fletcher, K., Warren, P. & Baines, D. (2005) Impact of nest visits by human observers on hatching success in Lapwings Vanellus vanellus: a field experiment. Bird Study, 52, 221-223. Fox, A. D., Jones, T. A., Singleton, R. & Agnew, A. D. Q. (1994) Food supply and the effects of recreational disturbance on the abundance and distribution of wintering Pochard on a gravel pit complex in southern Britain. Hydrobiologia. Gallet, S. & Roze, F. (2001) Resistance of Atlantic Heathlands to trampling in Brittany (France): influence of vegetation type, season and weather conditions. Biological Conservation, 97, 189-198. Gander, H. & Ingold, P. (1997) Reactions of male alpine chamois Rupicapra r rupicappa to hikers, joggers and mountainbikers. Biological Conservation, 79, 107-109. Gardner, S. & Usher, M. (1989) Insect abundance on burned and cut upland Calluna heath. The Entomologist, 108, 147-157. George, M. (1992) The land use, ecology and conservation of Broadland, Packard Publishing, Chichester. Ghandi, K., Spence, J., Langor, D. & Morgantini, L. (2001) Fire residuals as habitat reserves for epigaeic beetles (Coleoptera: Carabidae and Staphylinidae). Biological Conservation, 102, 131-141. Gheskiere, T., Vincx, M., Weslawski, J. M., Scapini, F., Degraer, S., (), & (2005) Meiofauna as descriptor of tourism-induced changes at sandy beaches. Marine Environmental Research, 60, 245-265. Gilbert, J. C. (2000) High soil phospherous availability and the restoration of species rich grassland. Silsoe: Cranfield University. Gilbert, O. L. (1989) The Ecology of Urban Habitats, Chapman & Hall, London. Gill, J. A. (1996) Habitat choice in wintering pink-footed geese:quantifying the constraints determining winter site use. Journal of Applied Ecology, 33, 884-892. Gill, J. A. (2007) Approaches to measuring the effects of human disturbance on birds. Ibis, 149, 9-14. Gill, J. A., Norris, K. & Sutherland, W. J. (2001a) The effects of disturbance on habitat use by black- tailed godwits Limosa limosa. Journal of Applied Ecology, 38, 846-856. Gill, J. A., Norris, K. & Sutherland, W. J. (2001b) Why behavioural responses may not reflect the population consequences of human disturbance. Biological Conservation, 97, 265 - 268. Gill, J. A., Sutherland, W. J. & Watkinson, A. R. (1996) A method to quantify the effects of human disturbance on animal populations. Journal Of Applied Ecology, 33, 786-792. Gimingham, C. (1985) Age-related interactions between Calluna vulgaris and phytophagus insects. Oikos, 44, 12-16. Gimingham, C. H. (1992) The lowland heath handbook. English Nature Science No. 8. English Nature, Peterborough. Glaves, D., Haycock, N., Costigan, P., Coulson, J., Marrs, R., Robertson, P. & Younger, J. (2005) DEFRA review of the Heather and Grass Burning Regulations and code. Science Panel assessment of the effects of burning on biodiversity, soils and hydrology. Report to DEFRA Conservation, Uplands and Rural Division, London. Gloaguen, J. (1993) Spatio-temporal patterns in post-burn succession on Brittany heathland. Journal of Vegetation Science, 4, 561-566. Gochfeld, M. (1983) Colony Site Selection by Least Terns: Physical Attributes of Sites. Colonial Waterbirds., 6, 205-213. Godefroid, S. & Koedam, N. (2004a) The impact of forest paths upon adjacent vegetation: effects of the path surfacing material on composition and soil compaction. Biological Conservation, 119, 405-419.

116

Godefroid, S. & Koedam, N. (2004b) Interspecific variation in soil compaction sensitivity among forest floor species. Biological Conservation, 119, 207-217. Godefroid, S., Massant, W., Weyembergh, G. & Koedam, N. (2003) Impact of fencing on the recovery of the ground flora of heavily eroded slopes of a deciduous forest. Environmental Management, 32, 62-76. Goldsmith, F. B. (1983) Ecological effects of visitors and restoration of damaged areas. Conservation in perspective (eds A.Warren & F. B. Goldsmith). Wiley, London. Goss-Custard, J. D., Triplet, P., Sueur, F. & West, A. D. (2006) Critical thresholds of disturbance by people and raptors in foraging wading birds. Biological Conservation, 127, 88-97. Goss-Custard, J. D. & Verboven, N. (1993) Disturbance and feeding shorebirds on the Exe estuary. Wader Study Group Bulletin, 68, 59-66. Gough, M. W. & Marrs, R. H. (1990) A comparison of soil fertility between semi-natural and agricultural plant communities: Implications for the creations of species-rich grassland on abondoned agricultural land. Biological Conservation, 51, 83-96. Grabherr, G. (1985) Damage to vegetation by recreation in the German and Austrian Alps. The ecological impacts of outdoor recreation on mountains areas in Europe and North America (eds N. G. Bayfiel, Barrow & G.C.). Recreational Ecology Research Report 9. Green, K. & Sanecki, G. (2006) Immediate and short term responses of bird and mammal assemblages to a subalpine wildfire in the Snowy Mountains, Australia. Austral Ecology, 31, 673-681. Green, R. & Higginbottom, K. (2001) The negative effects of wildlife tourism on wildlife. Wildlife Tourism Research Report Series: No. 5. Cooperative Research Centre for Sustainable Tourism. Grime, J. (1963) An ecological investigation of two plant communities in Coombesdale on the Derbyshire limestone. Journal of Ecology, 51, 391-402. Grime, J. P. (1979) Plant strategies and vegetational proccesses., Wiley, Chichester. Grubb, T. G., Robinson, W. L. & Bowerman, W. W. (2002) Effects of watercraft on bald eagles nesting in Voyageurs National Park, Minnesota. Wildlife Society Bulletin, 30, 156-161. Gutzwiller, K. J. & Riffell, S. K. (2008) Does repeated human intrusion alter use of wildland sites by red squirrels? multiyear experimental evidence. Journal of Mammalogy, 89, 374-380. Hall, C. (1996) An assessment of human impacts on the Castle bottom to Yateley Common SSSI, Hampshire. English Nature, Lyndhurst, Hampshire. Hall Cushman, J. & Meentemeyer, R. K. (2008) Multi-scale patterns of human activity and the incidence of an exotic forest pathogen. Journal of Ecology, 96, 766-776. Hamberg, L., Lehvävirta, S., Malmivaara-Lämsä, M., Rita, H. & Kotze, D. J. (2008) The effects of habitat edges and trampling on understorey vegetation in urban forests in Helsinki, Finland. Applied Vegetation Science, 11, 83-98. Hamilton, A. (2000) The characteristics and effects of management fire on blanket bog vegetation in North-West Scotland. University of Edinburgh, Edinburgh. Harper, M., Dietrich, C., Larimore, R. & Tessene, P. (2000) Effects of prescribed fire on prairie arthropods: an enclosed study. Natural areas, 20, 325-335. Harris, S. & Rayner, J. M. V. (1986) Urban Fox (Vulpes vulpes) Population Estimates and Habitat Requirements in Several British Cities Journal of Animal Ecology, 55, 575-591. Harris, S. & Yalden, D. W. (2008) Mammals of the British Isles, Mammal Society. Harrison, C. (1981) Recovery of lowland grassland and heathalnd in southern England from disturbance by seasonal trampling. Biological Conservation, 19. Hawke, C. & Jose, P. (1996) Reedbed management for commercial and wildlife interests. RSPB, Sandy. Haworth, P. & Thompson, D. (1990a) Factors assocaited with the breeding distribution of upland birds in the South Pennines, England. Journal of Applied Ecology, 27, 562-577.

117

Haworth, P. F. & Thompson, D. B. A. (1990b) Factors associated with the breeding distrubution of upland birds in the south Pennines, England. Journal of Applied Ecology. Hearn, K. & Gilbert, M. (1977) The effects of the 1976 drought on sites of nature conservation interest in England and Wales. Nature Conservancy Council, Banbury. Hegetschweiler, K. T. (2008) Ecological impact of fire place use in urban forests for visitor management. Philosophisch-Naturwissenschaftlichen. University of Basel, Basel. Herrando, S., Brotons, L., Del Amo, R. & Llacuna, S. (2002) Bird community succession after fire in a Mediterranean shrubland. Ardea, 90, 303-310. Herrando, S., Brotons, L. & Llacuna, S. (2003) Does fire increase the spatial heterogeneity of bird communities in Mediterranean landscapes? Ibis, 145, 307-317. Hewitt, D. E. (1973) Human pressures on soils in coastal areas. Welsh Soils Discussion Group Report No. 14, pp. 50-62. Higginbottom, K. (2004) Wildlife tourism : impacts, management and planning. Cooperative Research Centre for Sustainable Tourism. Hill, D., Hockin, D., Price, D., Tucker, G., Morris, R. & Treweek, J. (1997) Bird disturbance: Improving the quality and utility of disturbance research. Journal of Applied Ecology, 34, 275-288. Hill, D., Hockin, D., Price, D., Tucker, G., Morris, R. & Treweek, J. (1997 ) Bird disturbance: Improving the quality and utility of disturbance research. JOURNAL OF APPLIED ECOLOGY, 34, 275-288. Hirons, G. & Thomas, G. (1993) Disturbance on estuaries: RSPB nature reserve experience. Wader Study Group Bulletin, 68, 72-78. Hirst, R. A., Pywell, R. F., Marrs, R. H. & Putwain, P. D. (2005) The resilience of calcareous and mesotrophic grasslands following disturbance. Journal of Applied Ecology, 42, 498-506. Hobbs, R. (1984) Length of burning rotation and community composition in high level Calluna- Eriophorum bog in N England. Vegetatio, 57, 129-136. Hobbs, R., Currall, J. & Gimingham, C. (1983) The use of 'Thermocolor' pyometers in the study of heath fire behaviour. Journal of Ecology, 72, 241-250. Hobbs, R. & Gimingham, C. (1984a) Studies of fire in Scottish heathland communities I. Fire characteristics. Journal of Ecology, 72, 233-240. Hobbs, R. & Gimingham, C. (1984b) Studies of fire in Scottish heathland communities I: Fire characteristics. Journal of Ecology, 72, 233-240. Hobbs, R. J. & Gimingham, C. H. (1984c) Studies of fire in Scottish heathland communities II Post fire vegetation development. Journal of Ecology, 72, 585-610. Hockin, D., Ounsted, M., Gorman, M., Hill, D., Keller, V. & Barker, M. A. (1992) Examination of the effects of Disturbance on birds with reference to its importance in Ecological Assessments. Journal of Environmental Management, 36, 253-286. Holden, J. (2001) Recent reduction of frost in the North Pennines. Journal of Meteorology, 26, 369- 374. Holden, J. & Adamson, J. K. (2002) The Moor House long-term upland temperature record - new evidence of recent warming. Weather, 57, 119-127. Holden, J., Chapman, P., Evans, M., Hubacek, K., Kay, P. & Warburton, J. (2007) Vulnerability of organic soils in England and Wales. DEFRA and CCW. Holt, T. J., Rees, E. I., Hawkins, S. J. & Seed, R. (1998) Biogenic reefs (Volume IX). An overview of dynamic and sensitivity characteristics for conservation management of marine SACs. Scottish Association for Marine Science (UK Marine SACs Project). Howe, M. A., Hinde, D., Bennett, D. & Palmer, S. (2004) The conservation of the belted beauty Lycia zonaria britannica in the United Kingdom. Journal of Insect Conservation, 8, 159-166. Hunt, G. L. (1972) Influence of food distribution and human disturbance on the reproductive success of Herring Gulls. Ecology, 53, 1051 - 1061. Hylgaard, T. & Liddle, M. J. (1981) The Effect of Human Trampling on a Sand Dune Ecosystem Dominated by Empetrum nigrum. The Journal of Applied Ecology, 18, 559-569.

118

Irvine, K. (2005) Influence of Trampling Intensity versus Hydration State on Loss of Biomass from the Intertidal Rockweed, Fucus gardneri. Coastal Management, 33, 471-481. Jackson, D. L. (2000) Guidance on the interpretation of the Biodiversity Broad Habitat Classification (terrestrial and freshwater types): definitions and the relationships with other habitat classifications. JNCC Report. Joint Nature Conservation Committee, Peterborough. Jackson, M. J. & James, R. (1979) The Influence of Bait Digging on Cockle, Cerastoderma edule, Populations in North Norfolk. The Journal of Applied Ecology, 16, 671-679. Jacquet, K. & Prodon, R. (2009) Measuring the post-fire resilience of a bird-vegetation system: a 28- year study in a Mediterranean oak woodland. Oecologia, 161, 801-811. James, S. (1988) The post-fire environment and earthworm populations in tallgrass prairie. Ecology, 69, 476-483. Jenkins, C., Haas, M. E., Olson, A. & Ruesink, J. L. (2002) Impacts of Trampling on a Rocky Shoreline of San Juan Island, Washington, USA. Natural Areas Journal, 22, 260-269. JNCC (2007) Second report by the UK under Article 17 on the implementation of the Habitats Directive from January 2001 to December 2006. Joint Nature Conservation Committee, Peterborough. Johnson, S. (1995) Spider communities in the canopies of annually burned and long term un-burned Spartina pectinata wetlands. Environmental Entomology, 24, 832-834. Jónsdóttir, I. S. (1991) Effects of grazing on tiller size and population dynamics in a colonal sedge (Carex bigelowii). Oikos, 62, 177-88. Joy, J. (1995) Heathland management for the silver-studded blue butterfly. English Nature, Peterborough. Kaiser, M. J., Galanidi, M., Showler, D. A., Elliott, A. J., Caldow, R. W. G., Rees, E. I. S., Stillman, R. A. & Sutherland, W. J. (2006) Distribution and behaviour of Common Scoter Melanitta nigra relative to prey resources and environmental parameters. Ibis, 148, 110-128. Kay, M. F. & Gilchrist, H. G. (1998) Distraction displays made by female Common Eiders, Somateria mollissima borealis, in response to human disturbance. Canadian Field-Naturalist, 112, 529- 532. Keirle, I. & Stephens, M. (2004) Do walkers stay on footpaths? An observational study of Cwm Idwal in the Snowdonia National Park. Policies, methods and tools for visitor management – proceedings of the second International Conference on Monitoring and Management of Visitor Flows in Recreational and Protected Areas. Rovaniemi, Finland. Keller, V. (1991) Effects of human disturbance on Eider Duckling Somateria mollissima in an estuarine habitat in Scotland. Biological conservation, 58, 213 - 228. Kellomaki, S. (1973) Ground cover response to trampling in a spruce stand of Myrtillus type. Silva Fennica, 7, 96-113. Kellomaki, S. & Saastamoinen, V. L. (1975) Trampling tolerance of forest vegetation. Acta Fores. Fennica, 147, 5-19. Keough, M. & Quinn, G. (1991) Casualty and choice of measurements for detecting human impacts in marine environments. Australian Journal of Marine and Freshwater Research+, 42, 539- 554. Keough, M. J. & Quinn, G. P. (1998) Effects of Periodic Disturbances from Trampling on Rocky Intertidal Algal Beds. Ecological Applications, 8, 141-161. Kerbiriou, C., Le Viol, I., Robert, A., Porcher, A., Gourmelon, F. & Julliard, R. (2009) Tourism in protected areas can threaten wild populations: from individual response to population viability of the chough Pyrrhocorax pyrrhocorax %J Journal of Applied Ecology. Journal of Applied Ecology, 46, 657-665. Kerney, M. P. (1999) Atlas of the land and freshwater molluscs of Britain and Ireland, Harley Books, Colchester. Key, R. (2000) Bare ground and the conservation of invertebrates. British Wildlife, 11, 183-192.

119

Killeen, I. J. & Light, J. M. (2002) The status, distribution and ecology of Paludinella littorina (Dele Chiaje,1828) (Gastropoda: Assimineidae) in the British Isles. Journal of Conchology, 37. Kirby, J., Davidson, N., Giles, N., Owen, M. & Spray, C. (2004) Waterbirds and Wetland Recreation Handbook, Wildfowl and Wetlands Trust, Slimbridge, Gloustershire. Kirby, J. & Tantram, D. (1999a) Monitoring heathland fires in Dorset: Phase 1a. Meteorological triggers and the evaluation of fire danger. Terra Environmental Consultancy. Report to DEFRA, Northampton. Kirby, J. S., Clee, C. & Seager, V. (1993) Impact and extent of recreational disturbance to wader roosts on the Dee estuary: some preliminary results. Wader Study Group Bulletin, 68, 53-58. Kirby, J. S. & Tantram, D. A. S. (1999b) Monitoring heathland fires in Dorset: Phase 1. Kirby, P. (1992) Habitat management for invertebrates: a practical handbook., Royal Society for the Protection of Birds, Sandy. Kirby, P. (2001) Habitat management for invertebrates: a practical handbook., Royal Society for the Protection of Birds, Sandy. Klaassen, M., Bauer, S., Madsen, J. & Tombre, I. Modelling behavioural and fitness consequences of disturbance for geese along their spring flyway. Journal of Applied Ecology, 0, ???-??? Klinger, R., Kutilek, M. & Shellhammer, H. (1989) Population responses of black-tailed deer to prescribed burning. The Journal of Wildlife Management, 53, 863-871. Kuntz, K. L. & Larson, D. W. (2006) Influences of Microhabitat Constraints and Rock-Climbing Disturbance on Cliff-Face Vegetation Communities. Conservation Biology, 20, 821-832. Kuss, F. R. (1986) A review of major factors influencing plant responses to recreation impacts. Environmental Management, 10, 637-650. Kuss, F. R., Graefe, A. R. & Vaske, J. J. (1990) Visitor Impact Management: Volume One - A Review of Research. National Parks and Conservation Association, Washington DC. Kuss, F. R. & Hall, C. N. (1991) Ground flora trampling studies: five years after closure. Environmental Management, 15. Kuss, F. R. & Morgan, J. M. (1984) Using the USLE to estimate the physical carrying capacity of natural areas for outdoor recreation planning. 39, 383-387. Laegdsgaard, P. (2006) Ecology, disturbance and restoration of coastal saltmarsh in Australia: a review. Wetlands Ecology and Management, 14, 379-399. Lafferty, K. D. (2002) Human distrubance of shorebirds on California beaches. USGS. Lafferty, K. D., Goodman, D. & Sandoval, C. P. (2006) Restoration of Breeding by Snowy Plovers Following Protection from Disturbance. Biodiversity and Conservation, 15, 2217-2230. Lake, S. (2002) The role of livestock in the conservation of lowland heaths., PhD Southampton University. Lake, S., Bullock, J. M. & Hartley, S. E. (2001) Impacts of livestock grazing on lowland heathland in the UK., English Nature. (English Nature Research Reports No. 422), Peterborough. Lake, S. & Underhill-Day, J. (1999) Effects of grazing on heathland flora. International seminar on heathland management in north west Europe. Programme Life "Gestion des lands du nord ouest l'Europe", pp. 150-158. Bretagne Vivante/SEPNB. Lance, A. N., Baugh, I. D. & Love, J. A. (1989) Continued footpath widening in the Cairngorm Mountains, Scotland. Biological Conservation, 49, 201-14. Lawesson, J. E. (1998) Quantitative vegetation ecology in Denmark. Journal of Vegetation Science, 9, 891-896. Lawton, J. H. (1990) Red Grouse populations and moorland management. British Ecological Society Ecological Issues No. 2. British Ecological Society. Legg, C. J., Maltby, E. & Proctor, M. C. F. (1992) The ecology of severe moorland fire on the North York Moors: seed distribution and seedling establishment of Callluna vulgaris. Journal of Ecology, 80, 737-752. Leney, F. M. (1974) The ecological effects of public pressure on picnic sites.

120

Lerwill, J. K., Jones, G. A. P. & Penrose, R. S. (2003) Seals in Wales: Response to a possible Phocine Distemper Virus (PDV) outbreak. CCW contract report. CCW. Liddle, M. J. (1997) Recreation Ecology, Chapman & Hall, London. Liddle, M. J. & Chitty, L. D. (1981) The Nutrient-Budget of Horse Tracks on an English Lowland Heath. The Journal of Applied Ecology, 18, 841-848. Liddle, M. J. & Grieg-Smith, P. (1975) A Survey of Tracks and Paths in a Sand Dune Ecosystem. II. Vegetation. The Journal of Applied Ecology, 12, 909-930. Lidgard, D. C. (1996) The Effects of Human Disturbance on the Maternal Behaviour & Performance of Grey Seals at Donna Nook, Lincolnshire, UK. British Ecological Society. Liles, G. (2003) Otter Breeding Sites: Conservation and Management. Conserving Natura 2000 Rivers Conservation Techniques Series No. 5. English Nature, Peterborough. Liley, D. (1999) Predicting the consequences of human disturbance, predation and sea-level rise for Ringed Plover population size. School of Biological Sciences. University of East Anglia, Norwich. Liley, D. (2004) Human impacts on the Castle Bottom to Yateley Common and Hawley Commons SSSI, Hampshire. RSPB. Liley, D. (2007) Access to the Countryside and Bird Conservation: Priorities for Research 2007. Footprint Ecology / Natural England. Liley, D. (2008) Development and the North Norfolk Coast: scoping document on the issues relating to access. Footprint Ecology. Liley, D. & Clarke, R. T. (2002) The impact of human disturbance and urban development on key heathland bird species in Dorset. Sixth National Heathland Conference (eds J. C. Underhill- Day & D. Liley). RSPB, Bournemouth. Liley, D., Clarke, R. T., Mallord, J. W. & Bullock, J. M. (2006a) The effect of urban development and human disturbance on the distribution and abundance of nightjars on the Thames Basin and Dorset Heaths. Natural England / Footprint Ecology. Liley, D., Clarke, R. T., Underhill-Day, J. & Tyldesley, D. T. (2006b) Evidence to support the Appropriate Assessment of development plans and projects in south-east Dorset. Footprint Ecology / Dorset County Council. Liley, D., Langston, R. & Walker, J. (2002) Sustainable recreation management on lowland heathland Sites of Special Scientific Interest. RSPB . English Nature. Liley, D. & Sutherland, W. J. (2007) Predicting the population consequences of human disturbance for Ringed Plovers Charadrius hiaticula: a game theory approach. Ibis, 149, 82-94. Lindsay (2001). Littlemore, J. & Barker, S. (2001) The ecological response of forest ground flora and soils to experimental trampling in British urban woodlands. Urban Ecosystems, Volume 5, Number 4 / December, 2001, 257-276. Lloyd, P. (1968) The ecological significance of fire in limestone grassland communities of the Derbyshire Dales. Journal of Ecology, 56, 811-826. Lloyd, P. (1971) Effects of fire on the chemical status of herbaceous communities of the Derbyshire Dales. Journal of ecology, 59, 261-273. Locke, L. N., Kerr, S. M. & Zoromski, D. (1982) Lead Poisoning in Common Loons (Gavia immer). Avian Diseases, 26, 392-396. Lord, A., Waas, J. R., Innes, J. & Whittingham, M. J. (2001) Effects of human approaches to nests of northern New Zealand dotterels. Biological Conservation, 98, 233-240. Lowen, J., Liley, D., Underhill-Day, J. & Whitehouse, A. T. (2008) Access and Nature Conservation Reconciliation: supplementary guidance for England. Luff, M. L. (1998) Provisional atlas of the ground beetles (Coleoptera, Carabidae) of Britain, Biological Records Centre, Huntingdon. Lukac, G. & Hrsak, V. (2005) Influence of visitor numbers on breeding birds in the Paklenica National Park, Croatia. Ekologia-Bratislava, 24, 186-199.

121

Lutz, H. J. (1945) Soil conditions of picnic grounds in public forest parks. J. Forest, 43, 121-7. M., C. K. & J., S. W. (1999) A review of human disturbance effects on nesting colonial waterbirds. WATERBIRDS, 22, 68-79. Macdonald, A. (2000) Effects of grouse moor management on moorland plant diversity. Proceedings of the grouse moor management and biodiversity workshop. Napier University, Edinburgh. MacIntyre, M. (1991) The effects of walking and mountain bike use on the erosion of open heather moorland. Aberdeen Centre for Land Use, Aberdeen. Mader, H. J., Schell, C. & Kornacker, P. (1990) Linear barriers to arthropod movements in the landscape. Biological Conservation, 54, 209-222. Madsen, J. (1998) Experimental refuges for migratory waterfowl in Danish wetlands. I. Baseline assessment of the disturbance effects of recreational activities. Journal of Applied Ecology, 35, 386-397. Mallik, A. & Gimingham, C. (1985) Ecological effects of burning II. Effects on seed germination and vegetative regeneration. Journal of Ecology, 73, 633-644. Mallik, A., Gimingham, C. & Rahman, A. (1984) Ecological effects of heather burning I; Water infiltration, moisture retention and the porosity of the soil surface. Journal of Ecology, 72, 633-644. Mallik, A. & Rahman, A. (1985) Soil water repellancy in regularly burned Calluna heathlands: comparison of three measuring techniques. Journal of Environmental Management, 20, 207- 218. Mallord, J. W. (2005) Predicting the consequences of human disturbance, urbanisation and fragmentation for a woodlark Lullula arborea population. School of Biological Sciences. UEA, Norwich. Mallord, J. W., Dolman, P. M., Brown, A. F. & Sutherland, W. J. (2007) Linking recreational disturbance to population size in a ground-nesting passerine. Journal of Applied Ecology, 44, 185-195. Maltby, E., Legg, C. J. & Proctor, M. C. F. (1990) The ecology of severe fire on the North York Moors|: effects of the 1976 fires, and subsequent surface and vegetation development. Journal of Ecology, 79, 490-518. Manchester, S. J. & Bullock, J. M. (2000) The impacts of nonnative species on UK biodiversity and the effectiveness of control. Journal of Applied Ecology, 37, 845-864. Marrs, R. H. (1986) The role of catastrophic death of Calluna in heathland dynamics. Vegetatio, 66, 109-115. Marsden, S. J. (2000) Impact of Disturbance on Waterfowl Wintering in a UK Dockland Redevelopment Area. Environmental Management, 26, 207-213. Martin, J., de Neve, L., Fargallo, J. A., Polo, V. & Soler, M. (2004) Factors affecting the escape behaviour of juvenile chinstrap penguins, Pygoscelis antarctica, in response to human disturbance. Polar Biology, 27, 775-781. Mayer-Gross, H., Crick, H. Q. P. & Greenwood, J. J. D. (1997) The effect of observers visiting the nests of passerines: an experimental study. Bird Study, 44, 53-65. McGee, J. (1982) Small mammal populations in an unburned and early fire successional sagebrush community. Journal of Range Management, 35, 177-180. McMillan, M. A. & Larson, D. W. (2002) Effects of Rock Climbing on the Vegetation of the Niagara Escarpment in Southern Ontario, Canada. Conservation Biology, 16, 389-398. McQuaid-Cook, J. (1978) Effects of Hikers and Horses on Mountain Trails. Journal of Environmental Management, 6, 209-212. Medeirosa, R., Ramosa, J. A., Paivaa, V. H., Almeidac, A., Pedroa, P. & Antunes, S. (2007) Signage reduces the impact of human disturbance on little tern nesting success in Portugal. Biological Conservation, 135, 99-106. Merrett, P. (1976) Changes in the ground-living spider fauna after heathland fires in Dorset. Bulletin of the British Archaeological Society, 3, 214-221.

122

Mikola, J., Miettinen, M., Lehikoinen, E. & Lehtilae, K. (1994) The effects of disturbance caused by boating on survival and behaviour of velvet scoter Melanitta fusca ducklings. Biological Conservation, 67, 119-124. Milazzo, M., Badalamenti, F., Riggio, S. & Chemello, R. (2004) Patterns of algal recovery and small- scale effects of canopy removal as a result of human trampling on a Mediterranean rocky shallow community. Biological Conservation, 117, 191-202. Miller, G. & J., M. (1970) Regeneration of Calluna vulgaris (L) Hull at different ages and seasons in North-West Scotland. Journal of Applied Ecology, 7, 51-60. Miller, G. & Miles, J. (1984) Moorland management: A study of Exmoor. Institute of Terrestrial Ecology, Banchory. Milsom, T. P., Ennis, D. C., Haskell, D. J., Langton, S. D. & McKay, H. V. (1998) Design of grassland feeding areas for waders during winter: The relative importance of sward, landscape factors and human disturbance. Biological Conservation, 84, 119-129. Milwain, E. (1984) Recreational trampling pressure and its effects on the vegetation and soils of Les Quennevais - Blanches Banques Sand Dunes, St. Ouen's Bay, Jersey. Mitchell, B., Staines, B. W. & Welch, D. (1977) Ecology of red deer: a research review relevant to their management in Scotland. ITE. Moffett, M., McLachlan, A., Winter, P. & De Ruyck, A. (1998) Impact of trampling on sandy beach macrofauna. Journal of Coastal Conservation, 4, 87-90. Mohamed, B. F. & Gimingham, C. H. (1970) The morphology of vegetative regeneration in Calluna vulgaris. New Phytologist, 69, 743-750. Moneglia, P., Besnard, A., Thibault, J.-C. & Prodon, R. (2009) Habitat selection of the Corsican nuthatch Sitta whiteheadi after a fire. Journal of Ornithology, 150, 577-583. Monimeau, L., Mouillot, D., Fons, R., Prodon, R. & Marchand, B. (2002) Impact of prescribed burning on the survival rates of the wood mouse Apodemus sylvaticus. Acta Oecologica, 23, 51-58. Mooney, H. A. & Hobbs, R. J. (2000) Invasive Species in a Changing World, Island Press, Washington, DC. Mori, Y., Sodhi, N. S., Kawanishi, S. & Yamagishi, S. (2001) The effect of human disturbance and flock composition on the flight distances of waterfowl species. Journal of Ethology, 19, 115-119. Morris, P. A. & Harper, J. F. (1965) THE OCCURENCE OF SMALL MAMMALS IN DISCARDED BOTTLES. Proceedings of the Zoological Society of London, 145, 148-153. Moulton, N. & Corbett, K. (1999) The sand lizard conservation handbook, English Nature, Peterborough. Müller, S. W., Rusterholz, H.-P. & Baur, B. (2004) Rock climbing alters the vegetation of limestone cliffs in the northern Swiss Jura Mountains. Can. J. Bot, 82, 862-870. Murison, G. (2002) The impact of human disturbance on the breeding success of nightjar Caprimulgus europaeus on heathlands in south Dorset, England. English Nature, Peterborough. Murison, G. (2007) The impact of human disturbance, urbanisation and habitat type on a Dartford warbler Sylvia undata population. School of Biological Sciences. University of East Anglia, Norwich. Murison, G., Bullock, J. M., Underhill-Day, J., Langston, R., Brown, A. F. & Sutherland, W. J. (2007) Habitat type determines the effects of disturbance on the breeding productivity of the Dartford Warbler Sylvia undata. Ibis, 149, 16-26. Murray, S., Denis, T., Kido, J. & Smith, J. (1999) Human visitation and frequency and potential effects of collecting on rocky intertidal populations in southern California marine reserves. California Cooperative Oceanic Fisheries Investigations Reports, 40, 100-106. MV, B., Mountford, M., Abbott, A. & Vincent, S. (1976) The changes in ant species distribution during ten years post-fire regeneration of a heath. Journal of Animal ecology, 45, 115-133. Nature Conservancy Council (1983) The ecology and conservation of amphibian and reptile species endangered in Britain. Nature Conservancy Council, London.

123

Neal, E. G. (1977) Badgers, Blandford Press Ltd, Dorset. Neatherlin, E. A. & Marzluff, J. M. (2004) Responses of American crow populations to campgrounds in remote native forest landscapes. Journal of Wildlife Management, 68, 708–718. Newsome, D. (2003) The role of an accidentally introduced fungus in degrading the health of the Stirling Range National Park ecosystem in South Western Australia. Managing for Healthy Ecosystems (eds D. J. Rapport, W. L. lasley, D. E. Rolston, N. O. Nielsen, C. O. Qualset & A. B. Damania). Lewis Publishers, Boca Raton, Florida. Newsome, D., Cole, D. N. & Marion, J. L. (2004) Environmental impacts associated with recreational horse riding. (ed R. Buckley). CABI, Wallingford, Oxfordshire. Newsome, D., Milewski, A., Phillips, N. & Annear, R. (2002) Effects of Horse Riding on National Parks and Other Natural Ecosystems in Australia: Implications for Management. Journal of Ecotourism, 1, 52-74. Nimes, R. B. (1995) Bracken, heath and burning in the Quantock Hills, England. Heaths and moorlands: Cultural landscapes (eds D. B. A. Thomson, A. J. H. & M. B. Usher). HMSO, Edinburgh. Nimon, A. J., Schroter, R. C. & Oxenham, R. K. C. (1996) Artificial eggs: Measuring heart rate and effects of disturbance in nesting penguins. Physiology & Behavior, 60, 1019-1022. Nisbet, I. C. T. (2000a) Disturbance, habituation, and management of waterbird colonies - Commentary. waterbirds, 23, 312-332. Nisbet, I. C. T. (2000b) Disturbance, habituation, and management of waterbird colonies - Commentary waterbirds, 23, 312-332. Novoa, C., Dumas, S. & Prodon, R. (1998) Changes in reproductive habitat of gray partridge after burning. Journal of Range Management, 51, 607-613. Nuzzo, V. A. (1996) Structure of cliff vegetation on exposed cliffs and the effect of rock climbing. Canadian Journal of Botany, 74, 607–617. O'Connell, M. J., Ward, R. M., Jones, R., Onoufriou, C., Yallop, M. L., Winfield, I. J. & Harris, G. (2007) Effects of Recreational Disturbance on Waterbirds. Report to Natural England, by the Wildfowl and Wetlands Trust. Owens, N. W. (1977) Responses of wintering brent geese to human disturbance. Wildfowl, 28, 5-14. Panzer, R. (2002) Compatability of prescribed burning with the conservation of insects in small, isolated prairie reserves. Conservation Biology, 16, 1296-1307. Papouchis, C. M., Singer, F. J. & Sloan, W. B. (2001) Responses of desert bighorn sheep to increased human recreation. Journal of Wildlife Management, 65, 573-582. Peace, T. R. & Gilmour, J. S. L. (1949) The Effect of Picking on the Flowering of Bluebell, Scilla Non- Scripta. New Phytologist, 48, 115-117. Pearce-Higgins, J. W., Finney, S. K., Yalden, D. W. & Langston, R. H. W. (2007) Testing the effects of recreational disturbance on two upland breeding waders. Ibis, 149, 45-55. Pearce-Higgins, J. W. & Yalden, D. W. (1997) The effect of resurfacing the Pennine Way on recreational use of blanket bog in the Peak District national park, England. Biological Conservation, 82, 337 - 343. Pearce-Higgins, J. W. & Yalden, D. W. (2004) Habitat selection, diet, arthropod availability and growth of a moorland wader: the ecology of European Golden Plover Pluvialis apricaria chicks. Ibis, 146, 335-346. Pease, M. L., Rose, R. K. & Butler, M. J. (2005) Effects of human disturbances on the behavior of wintering ducks. Wildlife Society Bulletin, 33, 103-112. Penny Anderson Associates (2001a) Countryside and Rights of Way Act 2000, Part I - Access to the countryside guidance for statutory authorities involved in assessing the nature conservation implications of a statutory right of access in England and Wales under clause 26. . English Nature. Penny Anderson Associates (2001b) Countryside and Rights of Way Act 2000, Part I - Access to the countryside guidance for statutory authorities involved in assessing the nature conservation

124

implications of a statutory right of access in England and Wales under clause 26. English Nature. Penny Anderson Associates (2006) A Review of the Effects of Recreation and Sport on Nature Conservation. English Nature, Peterborough. Percival, S. M. (1993) The Effects of Reseeding, Fertilizer Application and Disturbance on the Use of Grasslands by Barnacle Geese, and the Implications for Refuge Management. Journal of Applied Ecology, 30, 437-443. Percival, S. M. & Evans, P. R. (1997) Brent Geese Branta bernicla and Zostera; Factors affecting the exploitation of a seasonally declining food resource. Ibis, 139, 121-128. Percival, S. M., Halpin, Y. & Houston, D. C. (1997) Managing the distribution of barnacle geese on Islay, Scotland, through deliberate human disturbance. Biological Conservation, 82, 273-277. Peters, K. A. & Otis, D. L. (2007) Shorebird roost-site selection at two temporal scales: is human disturbance a factor? Journal of Applied Ecology, 44, 196-209. Phillips, J., Yalden, D. & Tallis, J. (1981) Peak District moorland erosion study, Phase I report. Peak District Joint Planning Board, Bakewell. Pienkowski, M. J. (1984b) Breeding biology and population dynamics of Ringed Plovers Charadrius hiaticula in Britain and Greenland: nest predation as a possible factor limiting distribution and time of breeding. Journal of the Zoological society of London, 202, 83 - 114. Pierce, D. J. & Simons, T. R. (1986) The influence of human disturbance on Tufted Puffin breeding success. The Auk, 103, 214-216. Pinn, E. H. & Rodgers, M. (2005) The influence of visitors on intertidal biodiversity. Journal of the Marine Biological Association of the UK, 85, 263-268. Pons, P. (1998) Bird site tenacity after prescribed burning in a Mediterranean shrubland. Fire management and landscape ecology (ed L. Trabaud), pp. 261-270. International Association of Wildland Fire, Washington. Pons, P., Pierre-Yves, H., Gargallo, G., Prodon, R. & Lebreton, J.-D. (2003) Local survival after fire in Mediterranean shrublands: combining capture-recapture data over several bird species. Population Ecology, 45, 187-196. Poole, A. (2003) Analysis of chough productivity and recreational use in Pembrokeshire. CCW, Bangor. Prenda, J., Lopez-Nieves, P. & Bravo, R. (2001) Conservation of otter (Lutra lutra) in a Mediterranean area: the importance of habitat quality and temporal variation in water availability. Aquatic Conservation-Marine and Freshwater Ecosystems, 11, 343-355. Price, M. F. (1985) Impacts of Recreational Activities on Alpine Vegetation in Western North America. Mountain Research and Development, 5, 263-278. Prigioni, C. & Deflorian, M. C. (2005) Sett site selection by the Eurasian badger (Meles meles) in an Italian Alpine area. Italian Journal of Zoology, 72, 43-48. Randall, R. E. (2004) Management of coastal vegetated shingle in the United Kingdom. Journal of Coastal Conservation, 10, 159-168. Randler, C. (2006) Disturbances by dog barking increase vigilance in coots Fulica atra. European Journal of Wildlife Research, 52, 265-270. Ravenscroft, N., Parker, B., Vonk, R. & Wright, M. (2008) Disturbance to waterbirds wintering in the Stour-Orwell estuaries SPA. Suffolk Coasts and Heaths Unit. Rawes, M. & Hobbs, R. (1979) Management of semi-natural blanket bog in the Northern Pennines. Journal of Ecology, 67, 789-807. Read, H. (2000) Veteran trees, a guide to good management, English Nature, Peterborough. Rees, E. C., Bruce, J. H. & White, G. T. (2005) Factors affecting the behavioural responses of whooper swans (Cygnus c. cygnus) to various human activities. Biological Conservation, 121, 369-382. Rees, J. & Tivy, J. (1977) Recreational impact on lochshore vegetation. J. Scott. Assoc. Geogr. Teach., 6, 8-24. Rees, J. R. (1978) A 'people-counter' for unsurfaced wetland footpaths. Environ. Conserv., 5, 66-68.

125

Rees, S. L. (1990) The effects of trampling on grassland area, Brathens, Scotland an assessment of the relative impact of plimsols, walking boots and a mountain bike and a comparison of two different moisture levels. University of Leicester. Reijnen, R., Foppen, R. & Veenbaas, G. (1997) Disturbance by traffic of breeding birds: evaluation of the effect and considerations in planning and managing road corridors Biodiversity and Conservation, 6, 567-581. Remage-Healey, L. & Romero, L. M. (2000) Daily and seasonal variation in response to stress in captive starlings (Sturnus vulgaris): glucose. Gen Comp Endocrinol, 119, 60-8. Richardson, M. & Baines, D. (2004) An experimental assessment of the potential effects of human disturbance on black grouse in the north Pennines. Game Conservancy Trust / English Nature. Rickard, W. (1970) Ground dwelling beetles in burned and unburned vegetation. Journal of Range Management, 23, 293-297. Riddington, R., Hassel, M., Lane, S. J., Turner, P. A. & Walters, R. (1996) The impact of disturbance on the behaviour and energy budgets of Brent Geese. Bird Study, 43, 269 - 279. Ritchie, W. (2001) Coastal dunes: resultant dynamic position as a conservational managerial objective. Coastal Dune Management: Shared experience of European Conservation Practice. European symposium on coastal dunes of the Atlantic biogeographical region. Liverpool University Press, Southport. Robel, R., Hughes, J., Hull, S., Kemp, K. & Klute, D. (1998) Spring burning: resulting avian abundance and nesting in Kansas CRP. Journal of Range Management, 51, 132-138. Robert, H. C. & Ralph, J. C. (1975) Effects of Human Disturbance on the Breeding Success of Gulls. The Condor, 77, 495-499. Roberts, S. J., Lewis, J. M. S. & Williams, I. T. (1999) Breeding European Honey buzzards in Britain. British Birds, 92, 326-345. Robitaille, J.-F. & Laurence, S. (2002) Otter, Lutra lutra, occurrence in Europe and in France in relation to landscape characteristics. Animal Conservation, 5, 337-344. Rodway, M. S., Montevecchi, W. A. & Chardine, J. W. (1996) Effect of investigator disturbance on breeding success of Atlantic puffins Fratercula artica. Biological Conservation, 76, 311-319. Rodwell, J. S. (1991) British Plant Communities, Volume 2: Mires and heath, Cambridge University Press, Cambridge. Rogova, T. V. (1976) Influence of trampling on vegetation of forest meadows and whortleberry- moss-pine forest cenoses. Sov. J. Ecol., 7, 356-359. Ronconi, R. A. & St. Clair, C. C. (2002) Management options to reduce boat disturbance on foraging black guillemots (Cepphus grylle) in the Bay of Fundy. Biological Conservation, 108, 265-271. Root, J. D. a. K., L.J., . (1972) Trail conditions along a portion of the Great Divide trail route, Alberta and British Colombia Mountains. Research Council of Alberta,, Edmonton, Alberta. Roovers, P., Verheyen, K., Hermy, M. & Gulinck, H. (2004) Experimental trampling and vegetation recovery in some forest and heathland communities. Applied Vegetation Science, 7, 111-118. Rose, R. & Clarke, R. (2005) Urban impacts on Dorset heathlands: Analysis of the heathland questionnaire survey and heathland fires incidence data sets. Report No. 624. English Nature, Peterborough. Rose, R. J., Webb, N. R., Clarke, R. T. & Traynor, C. H. (2000) Changes on the heathlands in Dorset, England, between 1987 and 1996. Biological Conservation, 93, 117-125. Ross, S., Adamson, H. & Moon, A. (2003) Evaluating management techniques for controlling Molinia caerulea and enhancing Calluna vulgaris on upland wet heathland in Northern England, UK. Agriculture Ecosystems & Environment, 97, 39-49. Rossi, F., Forster, R. M., Montserrat, F., Ponti, M., Terlizzi, A., Ysebaert, T. & Middelburg, J. (2007) Human trampling as short-term disturbance on intertidal mudflats: effects on macrofauna biodiversity and population dynamics of bivalves. Marine Biology, 151, 2077-2090.

126

Rowell, T. A. (1988) The peatland management handbook. Research and Survey in Nature Conservation No 14. Nature Conservancy Council, Peterborough. Rowley, I. & Brooker, M. (1987) The response of a small insectiverous bird to fire in heathlands. Nature Conservation: the role of remnants of native vegetation (ed G. W. A. D. A. Saunders, A. Burbridge,& A. J. M. Hopkins). Surrey Beatty, Sydney. Rowley, I., Brooker, M. & Russell, E. (1991) The breeding biology of the splendid fairy-wren Malurus splendens: the significance of multiple broods. Emu, 91, 197-221. RSPB (1995) Consevation management of blanket bog. Scottish raised bog conservation project. Scottish Wildlife Trust, Edinburgh. Ruhlen, T. D., Abbott, S., Stenzel, L. E. & Page, G. W. (2003) Evidence that human disturbance reduces Snowy Plover chick survival. Journal of Field Ornithology, 74, 300-304. Rusterholz, H.-P., Müller, S. W. & Baur, B. (2004) Effects of rock climbing on plant communities on exposed limestone cliffs in the Swiss Jura mountains. Applied Vegetation Science, 7, 35-40. Sandvik H & RT, B. (2001) Effect of investigator disturbance on the breeding success of the Black- legged Kittiwake. JOURNAL OF FIELD ORNITHOLOGY, 72, 30-42. Satchell, J. E. & Marren, P. R. (1976) The effects of recreation on the ecology of natural landscapes. Nature and environment series No. 11. Council of Europe, Strasbourg. Saunders, C., Selwyn, J., Richardson, S., May, V. & Heeps, C. (2000) A review of the effects of recreational interactions within UK European marine sites. UK CEED & Bournemouth University. Schiel, D. & Taylor, D. (1999) Effects of trampling on a rocky intertidal algal assemblage in southern New Zealand. Journal of Experimental Marine Biology and Ecology, 235, 213-235. Schulz, R. & Stock, M. (1993) Kentish Plovers and tourists - competitors on sandy coasts ? Wader Study Group Bulletin, Special Issue: Davidson & Rothwell (eds.) Disturbance to Waterfowl on Esturaries, 68. Shardlow, M. E. (2001) A review of the Conservation Importance of Shingle Habitats for Invertebrates in the United Kingdom (UK). Ecology and Geomorphology of Coastal Shingle (eds J. R. Packham, R. E. Randall, B. R. S. K. & A. Neal). Westbury Academic and Scientific Publishing. Shaw, P. J. A., Lankey, K. & Hollingham, S. A. (1995) Impacts of trampling and dog fouling on vegetation and soil conditions on Headley Heath. The London Naturalist, 74, 77-82. Shaw, S., Wheeler, B., Kirby, P., Phillipson, P. & Edmunds, R. (1996) Literature review of the historical effects of burning and grazing of blanket bog and upland wet heath. English Nature Research Report 172. English Nature & Countryside Council for Wales, Peterborough. Sidaway, R. & Ramblers' Association. (1990) Birds and walkers : a review of existing research on access to the countryside and disturbance to birds, Ramblers' Association, London. Siikamäki, P., Törn, T. & Tolvanen, A. (2006) Environmental Impacts of Recreational Horse Riding in Protected Areas. Exploring the Nature of Management. Proceedings of the Third International Conference on Monitoring and Management of Visitor Flows in Recreational and Protected Areas. (eds D. Siegrist, C. Clivaz, M. Hunziker & S. Iten). University of Applied Sciences, Rapperswil, Rapperswil, Switzerland. Silbernagel, J. (2008) Human visitation in the rocky intertidal and its effects on abundance of algal and invertebrate species. Oregon state university, Corvallis, USA. Simmons, I. (2003) Moorlands of England and Wales - An Environmental History 8,000 BC - AD 2,000, Edinburgh University Press, Edinburgh. Simms, C. (1969) Recolonisation of burnt heath by lizards. British journal of Herpetology. Skeate, E. & Perrow, M. (2008) An undercurrent of change for Norfolk's seals. British Wildlife, 19, 305-314. Slatter, R. J. (1978) Ecological effects of trampling on sand dune vegetation. J. of Biological Education, 12, 89-96.

127

Smit, C. J. & Visser, G. J. M. (1993) Effects of disturbance on shorebirds: a summary of existing knowledge from the Dutch Wadden Sea and Delta area. Wader Study Group Bulletin, 68, 6- 19. Smith, A., Redpath, S., Campbell, S. & Thirgood, S. (2001) Meadow pipits, red grouse and the habitat characteristics of managed grouse moors. Journal of Applied ecology, 38, 390-400. Smith, J. R., Fong, P. & Ambrose, R. F. (2008) The Impacts of Human Visitation on Mussel Bed Communities Along the California Coast: Are Regulatory Marine Reserves Effective in Protecting These Communities? Environmental Management, Volume 41, 599-612. Sneddon, P. & Randall, R. E. (1989) Coastal vegetated shingle structures of Great Britain: bibliography. Research and survey in nature conservation. Nature Conservancy Council, Peterborough. Sneddon, P. & Randall, R. E. (1993a) Coastal vegetated shingle structures of Great Britain: Appendix 1 - Wales. Joint Nature Conservation Committee, Peterborough. Sneddon, P. & Randall, R. E. (1993b) Coastal vegetation shingle structures of Great Britain: Main Report. Joint Nature Conservation Committee, Peterborough. Sneddon, P. & Randall, R. E. (1994) Coastal vegetated shingle structures of Great Britain: Appendix 3 - England. Joint Nature Conservation Committee, Peterborough. Speight, M. (1973) Outdoor Recreation and its Ecological Effects: a bibliography and review. Discussion Papers in Conservation No. 4, pp. 35. University College, London. Spellerberg, I. (1977) Behaviour of a yound smooth snake Coronella austriaca Laurenti. Biological Journal of the Linnean Society, 9, 323-330. Spokes, K. E. (1997) An investigation into the effects of trampling on shingle vegetation. No place of publication given. Stalmaster, M. V. & Kaiser, J. L. (1997) Flushing responses of wintering bald eagles to military activity. Journal Of Wildlife Management, 61, 1307-1313. Stammel, B. & Kiehl, K. (2004) Do hoof prints actually serve as a regeneration niche for plant species in fens? Phytocoenologia, 34, 271-286. Stankey, G. H. & Manning, R. (1986) Carrying capacity of recreation settings. President's Commission on Americans Outdoors: a literature review. U.S Government Printing Office, Washington DC. Stevens, M. (2007) The impact of severe fire on small mammals. Deakin University, Melbourne, Australia. Stevenson, A., Rhodes, A., Kirkpatrick, A. & Macdonald, A. (1996) The determination of fire histories and an assessment of their effects on moorland soils and their vegetation. Survey and monitoring report No 16. Scottish Natural Heritage, Edinburgh. Stewart, A., Pearman, D. A. & Preston, C. D. (1994) Scarce Plants in Britain, JNCC. Stewart, G., Coles, C. & Pullin, A. (2004a) Does burning degrade blanket bog? Systematic Review No. 1. Centre for Evidence Based Conservation, University of Birmingham, Birmingham. Stewart, G. B., Coles, C. F. & Pullin, A. S. (2004b) Does burning of UK sub-montane, dry dwarf scrub heath maintain vegetation diversity? Systematic Review No. 2. Centre for Evidence based Conservation. Birmingham University, Birmingham. Stillman, R. A. & Goss-Custard, J. D. (2002a) Seasonal changes in the response of oystercatchers Haematopus ostralegus to human disturbance. Journal of Avian Biology, 33, 358-365. Stillman, R. A. & Goss-Custard, J. D. (2002b) Seasonal changes in the response of oystercatchers Haematopus ostralgus to human disturbance. Journal of Avian Biology, 33, 358 - 365. Stillman, R. A., Goss-Custard, J. D., West, A. D., Durell, S., McGrorty, S., Caldow, R. W. G., Norris, K. J., Johnstone, I. G., Ens, B. J., Van der Meer, J. & Triplet, P. (2001) Predicting shorebird mortality and population size under different regimes of shellfishery management. Journal of Applied Ecology, 38, 857-868. Stillman, R. A., West, A. D., Caldow, R. W. G. & Durell, S. E. A. L. V. D. (2007) Predicting the effect of disturbance on coastal birds. Ibis, 149, 73-81.

128

Stock, M. (1993) Studies on the effects of disturbance on staging brent geese: a progress report. Wader Study Group Bulletin, 68, 29-34. Stock, M. & Hofeditz, F. (1997) Compensatory limits: energy budgets of Brent Geese, Branta b- bernicla, the influence of human disturbance. Journal Fur Ornithologie, 138, 387-411. Stone, E. L., Jones, G. & Harris, S. (2009) Street lighting disturbs commuting bats. Current Biology, 19, 1123-1127. Strauss, E. G. (1990) Reproductive success, life history patterns, and behavioural variation in a population of Piping Plover subjected to human disturbance. Tufts. Streeter, D. T. (1971) The effects of public pressure on the vegetation of chalk downland at Box Hill, Surrey. The Scientific Management of Animal and Plant Communities for Conservation (eds E. Duffey & A.S. Watt), pp. 459-468. Blackwell, Oxford. Stubbs, A. E. & Drake, M. (2001) British Soldierflies and their Allies. BENHS. Studlar, S. M. (1980) Trampling effects on bryophytes: trail surveys and experiments. Bryol., 83, 301- 313. Sukopp, H. (1971) Effects of man, especially recreational activities, on littoral macrophytes. Hydrobiologia, 12, 331-340. Summer, R. M. (1986) Geomorphic impacts of horse traffic on montane land forms. Journal of Soil and Water Conservation, 41, 126-128. Sutherland, W. J. (2007) Future directions in disturbance research. Ibis, 149, 120-124. Symes, N. & Day, J. (2003) A practical guide to the restoration and management of lowland heathland, RSPB, Sandy, Beds. Tallis, J. (1973) Studies on southern Pennine peats. V. Direct observations on peat erosion and peat hydrology at Featherbed Moss, Derbyshire. Journal of Ecology, 61, 1-22. Tallis, J. (1987) Fire and flood at Holme Moss: Erosion processes in an upland balnket mire. Journal of Ecology, 75, 1099-1129. Tantram, D., Boobyer, M. & Kirby, J. (1999) Monitoring heathland fires in Dorset: Phase 2. Report to DETR, Northampton. Taylor, A. R. & Knight, R. L. (2003) Wildlife responses to recreation and associated visitor perceptions. Ecological Applications, 13, 951-963. Taylor, E. C., Green, R. E. & Perrins, J. (2007) Stone-curlews Burhinus oedicnemus and recreational disturbance: developing a management tool for access. Ibis, 149, 37-44. Taylor, K., Anderson, P., Liley, D. & Underhill-Day, J. C. (2006) Promoting Positive Access Management to Sites of Nature Conservation Value: A Guide to Good Practice, English Nature / Countryside Agency, Peterborogh and Cheltenham. Taylor, K., Anderson, P., Taylor, R. P., Longden, K. & Fisher, P. (2005) Dogs, access and nature conservation. Research Report. English Nature, Peterborough. Telfer, M. G. (2006) Invertebrate survey of the soft-rock cliffs of Norfolk. Buglife, Peterborough. Telfer, M. G. & Eversham, B. C. (2000) Species Recovery Programme: workshop on the ecology and conservation of BAP ground-beetles. Unpublished report to English Nature. Tempel, D. J. & Gutierrez, R. J. (2003) Fecal corticosterone levels in California spotted owls exposed to low-intensity chainsaw sound. Wildlife Society Bulletin, 31, 698-702. Tharme, A. P., Green, R. E., Baines, D., Bainbridge, I. P. & Brien, M. (2001) The effect of management for red grouse shooting on the population density of breeding birds on heather-dominated moorland. Journal of Applied Ecology, 38, 439-457. Thayer, J. A., Sydeman, W. J., Fairman, N. P. & Allen, S. G. (1999) Attendance and Effects of Disturbance on Coastal Common Murre Colonies at Point Reyes, California. Waterbirds, 22, 130-139. The Mountaineering Council of Scotland (Undated) Memorials (http://www.mcofs.org.uk/memorials.asp). The Nevis Partnership (2009) Memorials (http://www.nevispartnership.co.uk/policy.asp).

129

Thomas, K., Kvitek, R. G. & Bretz, C. (2003) Effects of human activity on the foraging behavior of sanderlings Calidris alba. Biological Conservation, 109, 67-71. Thomas, R. (1991) The ecological effects of paintball games on ancient semi-natural woodland. Belham Wood Final Report. English Nature. Thomas, R., Anderson, P. & Radford E. (1994) The ecological effects of woodland recreation. Quarterly J. Forestry, 88, 225-232. Thurston, E. & Reader, R. J. (2001) Impacts of Experimentally Applied Mountain Biking and Hiking on Vegetation and Soil of a Deciduous Forest. Environmental Management, 27, 397-409. Toro, M. & Granados, I. (2002) Restoration of a Small High Mountain Lake after Recent Tourist Impact: The Importance of Limnological Monitoring and Palaeolimnology. Water, Air, & Soil Pollution: Focus, 2, 295-310. Townshend, D. J. & O'Connor, D. A. (1993) Some effects of disturbance to waterfowl from bait digging and wildfowling at Lindisfarne National Nature Reserve, north-east England. Wader Study Group Bulletin, 68, 47-52. Tratalos, J. A., Gill, J. A., Bateman, I., Watkinson, A. R. & Sutherland, W. J. (2005) Interactions between tourism, breeding birds and climate change across a regional scale. Tyndall Centre for Climate Change Research, Norwich. Trollope, W. (1984) Fire behaviour, Springer verlag, New York. Tucker, G. (2003) Review of the impacts of heather and grassland burning in the upland on soils, hydrology and biodiversity. English Nature, Peterborough. Turner, M. G. (1987) Effects of grazing by feral horses, clipping, trampling, and burning on a Georgia salt marsh. Estuaries, 10, 54-60. Turpie, J. K. (1995) Non-breeding territoriality: causes and consequences of seasonal and individual variation in Grey Plover Pluvialis squatarola. Journal of Animal Ecology, 64, 429-438. Tyler-Walters, H. & Arnold, C. (2008a) Sensitivity of Intertidal Benthic Habitats to Impacts Caused by Access to Fishing Grounds. CCW Policy Research Report. CCW, Bangor. Tyler-Walters, H. & Arnold, C. (2008b) Sensitivity of Intertidal Benthic Habitats to Impacts Caused by Access to Fishing Grounds. Report to Cyngor Cefn Gwlad Cymru / Countryside Council for Wales from the Marine Life Information Network (MarLIN). Marine Biological Association of the UK, Plymouth. UK Biodiversity Group (1999a) Species Action Plan for Formica rufibarbis. Tranche 2 Action Plans - Volume IV: Invertebrates. UK Biodiversity group. UK Biodiversity Group (1999b) Species Action Plan for Thyridanthrax fenestratus. Tranche 2 Action Plans - Volume IV: Invertebrates. UK Biodiversity group. Underhill-Day, J. (2005) A literature review of urban effects on lowland heaths and their wildlife. English Nature Research Report No. 623. English Nature, Peterborough. Urfi, A. J., Goss-Custard, J. D. & Lev. Dit Durell, S. E. A. (1996) The Ability of Oystercatchers Haematopus ostralegus to Compensate for Lost Feeding Time: Field Studies on Individually Marked Birds. Journal of Applied Ecology, 33, 873-883. Usher, M. & Smart, L. (1988) Recolinisation of burnt and cut heathland in the North York Moors by arachnids. The Naturalist, 113, 103-111. Usher, M. B. (1992) Management and Diversity of Arthropods in Calluna Heathland. Biodiversity and Conservation, 1, 63-79. V., B. M., Mountford, M. D., Abbott, A. & Vincent, S. (1976) The changes in ant species distribution during ten years post-fire regeneration of a heath. Journal of Animal ecology, 45, 115-133. Valette, J., Gomendy, V., Marechal, J., Houssard, C. & Gillon, D. (1994) Heat transfer in the soil during low-intensity experimental fires:the role of duff and soil moisture content. International Journal of Wilfire, 4, 225-237. Van de Werfhorst, L. & Pearse, J. (2007) Trampling in the rocky intertidal of central California: A follow-up study. Bulletin of Marine Science, 81, 245-254. Van der Zande, A. N. (1984) Outdoor recreation and birds: conflict or symbiosis. Leiden.

130

Van der Zande, A. N., Berkhuizen, J. C., Latesteijn, H. C. v., Keurs, W. J. t. & Poppelaars, A. J. (1984) Impact of outdoor recreation on the density of a number of breeding bird species in woods adjacent to urban residential areas. Biological Conservation, 30, 1-39. Van Der Zande, A. N. & Verstrael, T. J. (1985) Impacts of outdoor recreation upon nest site choice and breeding success of the Kestrel. Ardea, 73, 90-99. Verboven, N., Ens, B. J. & Dechesne, S. (2001) Effect of Investigator Disturbance on Nest Attendance and Egg Predation in Eurasian Oystercatchers. Auk, 118, 503-508. Verhulst, S., Oosterbeek, K. & Ens, B. J. (2001) Experimental evidence for effects of human disturbance on foraging and parental care in oystercatchers. Biological Conservation, 101, 375-380. Vujnovic, K., Wein, R. W. & Dale, M. R. T. (2002) Predicting plant species diversity in response to disturbance magnitude in grassland remnants of central Alberta. Can. J. Bot, 80, 504-511. Walker, B. G., Dee Boersma, P. & Wingfield, J. C. Habituation of Adult Magellanic Penguins to Human Visitation as Expressed through Behavior and Corticosterone Secretion. Conservation Biology, 0. Warwick, S. I. & Briggs, D. (1978) The genecology of lawn weeds. Evidence for disruptive selection in Poa annua L. in a mosaic environment of bowling greens, lawns and flower beds. New Phytologist, 81, 725-37. Warwick, S. I. & Briggs, D. (1979) The genecology of lawn weeds. III. Cultivation experiments with Achillea millefolium L., Bellis perennis L., Plantago lanceolata L., Plantago major L., and Prunella vulgaris L. collected from lawns and contrasting grassland habitats. New Phytologist, 83, 509-36. Watson, A. (1979) Bird and mammal numbers in relation to human impact at ski lifts on Scottish Hills. Journal of Applied Ecology, 16. Watson, A., Miller, G. R. & Green, F. H. W. (1966) Winter browning of heather (Calluna vulgaris) and other moorland plants. Trans. Bot. Soc. Edinb., 40, 195-203. Weaver, T. & Dale, D. (1978) Trampling Effects of Hikers, Motorcycles and Horses in Meadows and Forests. The Journal of Applied Ecology, 15, 451-457. Weaver, T., Dale, D. & Hartley, E. (1979) The Relationship of Trail Condition to Use, Vegetation, User, Slope, Season and Time. Recreational Impacts on Wildlands Conference (eds R. Ittner, D. R. Potter, J. K. Agee & S. Anschell), pp. 94-100. Seattle, Washington. Webb, N. (1997) The effects of fires on the dynamics of heathland communities in Southern England. NNA-Berechte, 10, 134-141. Webb, N. R. (1986) Heathlands, Heathland Changes, Collins, London. Webb, N. R. (1990) Changes in the heathlands of Dorset, England, between 1978 and 1987. Biological Conservation, 51, 273-286. Webb, N. R. & Haskins, L. E. (1980) An ecological survey of heathlands in the Poole Basin, Dorset, England, in 1978. Biological Conservation, 17, 281-296. Webb, N. V. & Blumstein, D. T. (2005) Variation in human disturbance differentially affects predation risk assessment in Western Gulls. Condor, 107, 178-181. Webster, S. & Adams, N. (1989) The ecological effects of paintball games on an ancient woodland. The Planner, 75, 11-13. Weimerskirch, H., Shaffer, S. A., Mabille, G., Martin, J., Boutard, O. & Rouanet, J. L. (2002) Heart rate and energy expenditure of incubating wandering albatrosses: basal levels, natural variation, and the effects of human disturbance. J Exp Biol, 205, 475-83. West, A. D., Goss-Custard, J. D., Stillman, R. A., Caldow, R. W. G., Durell, S. & McGrorty, S. (2002) Predicting the impacts of disturbance on shorebird mortality using a behaviour-based model. Biological Conservation, 106, 319-328. West, A. D., Yates, M. G., McGrorty, S. & Stillman, R. A. (2007) Predicting site quality for shorebird communities: A case study on the Wash embayment, UK. Ecological Modelling, 202, 527- 539.

131

Westcott, S. M. & Stringell, T. B. (2003) Field observations on the disturbance of grey seals in North Wales, 2003. CCW marine monitoring reports. Weston, M. A. & Elgar, M. A. (2005) Disturbance to brood-rearing Hooded Plover Thinornis rubricollis: responses and consequences. Bird Conservation International, 15, 193-209. Whinam, J. & Chilcott, N. M. (2002) Impacts after four years of experimental trampling on alpine/sub-alpine environments in western Tasmania. Journal of Environmental Management, 67, 339-351. Whitehouse, A. T. (2007) Managing coastal soft cliffs for invertebrates. Buglife, Peterborough. Whitfield, D. P., Ruddock, M. & Bullman, R. (2008) Expert opinion as a tool for quantifying bird tolerance to human disturbance. Biological Conservation, 141, 2708-2717. Whittaker, C. (1961) Temperatures in heath fires. Journal of Ecology, 49, 709-715. Whittaker, E. & Gimingham, C. (1962) The effect of fire on the regeneration of Calluna vulgaris (L) Hull from seed. Journal of Ecology, 50, 815-882. Wichmann, M. C., Alexander, M. J., Soons, M. B., Galsworthy, S., Dunne, L., Gould, R., Fairfax, C., Niggemann, M., Hails, R. S. & Bullock, J. M. (2009) Human-mediated dispersal of seeds over long distances. 276, 523-532. Wilson, J. P. & Seney, J. P. (1994) Erosional Impact of Hikers, Horses, Motorcycles, and Off-Road Bicycles on Mountain Trails in Montana. Mountain Research and Development, 14, 77-88. Wingfield, J. C. & Sapolsky, R. M. (2003) Reproduction and resistance to stress: When and how. Journal of Neuroendocrinology, 15, 711-724. Wolton, R. J. (1991) Management guidelines for culm grassland (rhos pasture in south-west England). English Nature, Okehampton. Woodfield, E. & Langston, R. (2004) Literature review on the impact on bird populations of disturbance due to human access on foot, Royal Society for the Protection of Birds, Sandy, Beds. Wynberg, R. P. & Branch, G. M. (1994) Disturbance associated with bait-collection for sandprawns (Callianassa kraussi) and mudprawns (Upogebia africana): Long-term effects on the biota of intertidal sandflats. Journal of Marine Research, 52, 523-558. Yalden, D. W. (1990) Communications from the Mammal Society – No. 60: Recreational disturbance of large mammals in the Peak District. Journal of Zoological Society of London, 221, 293-326. Yalden, D. W. (1992) The influence of recreational disturbance on Common Sandpipers Actitis hypoleucos breeding by an upland reservoir in England. Biological Conservation, 61, 41 - 49. Yalden, D. W. & Yalden, P. E. (1988) The level of recreational pressure on blanket bog in the Peak District national park, England. Biological Conservation, 44, 213 - 227. Yalden, D. W., Yalden, P. E. & Nature Conservancy Council. (1989) Golden plovers and recreational disturbance, Nature Conservancy Council, Edinburgh. Yalden, P. E. & Yalden, D. W. (1990) Recreational disturbance of breeding Golden Plovers Pluvialis apricarius. Biological conservation, 51, 243 - 262. Yasué, M. (2005) The effects of human presence, flock size and prey density on shorebird foraging rates. Journal of Ethology, 23, 199-204. Yasue, M. & Dearden, P. (2006) The potential impact of tourism development on habitat availability and productivity of Malaysian plovers Charadrius peronii. Journal of Applied Ecology, 43, 978-989. Yeo, M. (1997) Blanket mire degradation in Wales. Blanket mire degradation: Causes, consequences and challenges (ed R. M. P. D. H. J. H. Tallis). University of Manchester. Young, J., Halada, L., Kull, T., Kuzniar, A., Tartes, U., Uzunov, Y. & Watt, A. e. (2004) Conflicts between human activities and the conservation of biodiveresity in agricultural landscapes, grasslands, forests, wetlands and uplands in the Acceeding and Candidate Countries: a report of the BIOFORUM project, March 2004. Centre for Ecology & Hydrology, Banchory, UK. Zwolak, R. & Foresman, K. (2007) Effects of a stand-replacing fire on small-mammal communities in montane forest. Canadian Journal of Zoology, 85, 815-822.

132

Appendix I: Welsh bird species potentially vulnerable to disturbance

The following list is based on that produced by the Welsh Ornithological Society (WOS) and published on their website13. Only species that are listed as birds of conservation concern in the UK are included (Eaton et al., 2009), and we have also excluded species that are particularly rare or scarce in Wales (i.e. species that are listed by WOS as former breeders, casual breeders, scarce visitors or rare visitors). We highlight those species listed by Natural England as potentially vulnerable to disturbance - their “Category A” list (as set out in Lowen et al., 2008).

Key:

Status

RB Resident breeder Present all year and breeds MB Migrant breeder Summer visitor and breeds IB1 Species introduced pre 1950 Introduced resident breeder IB2 Species introduced post 1950 Introduced resident breeder CB1 Casual breeder - pre 1950 Has bred on <5 occasions CB2 Casual breeder - post 1950 Has bred on <5 occasions FB Former breeder Has bred both pre an dpost 1950 FB1 Former Breeder -pre 1950 Has bred, but never became established FB2 Former Breeder - post 1950 Has bred, but never became established WV Winter visitor Present during winter months PV Passage visitor Spring and autumn passage migrant RV3 Rare visitor - BBRC Vagrant visitor RV4 Rare visitor - WRP Rare visitor (<5 records per year SV Scarce visitor Scarce visitor (5-20 records per year) E Escape Escape RN Resident non-breeder Present

BOCC

R Red listed A Amber listed

13 http://www.welshos.org.uk/species.htm

133

Species list

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Bewick's Swan Cygnus bewickii Alarch Bewick WV/PV A Little direct evidence of disturbance but potentially similar to Whooper Swan. Alarch y Whooper Swan Cygnus cygnus WV A Some evidence of behavioural response (Rees et al., 2005) Gogledd Pink-footed Anser Gŵydd droed- Evidence for disturbance effects on both staging and wintering birds (Klaassen et al., Gill et al., 1996, Gill, WV A Goose brachyrhynchus binc 1996) A range of studies, mostly relating to agricultural systems (Cope et al., 2003, Percival, 1993, Percival et al., Barnacle Goose Branta leucopsis Gŵydd Wyran WV A 1997) A number of studies showing range of impacts (Milsom et al., 1998, Owens, 1977, Percival and Evans, Brent Goose Branta bernicla Gŵydd Ddu WV/PV A 1997, Riddington, 1996, Stock, 1993, Stock and Hofeditz, 1997) Hwyaden yr Densities on estuaries in winter shown to be lower close to footpaths (Burton et al., 2002b). Nesting birds Shelduck Tadorna tadorna RB/WV A Eithin prone to desert if disturbed (Crick et al., 2003). Literature includes evidence of impacts from bait digging and angling (Cryer et al., 1987, Madsen, 1998, Wigeon Anas penelope Chwiwell WV/CB A Pease et al., 2005, Townshend and O'Connor, 1993) A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Gadwall Anas strepera Hwyaden Lwyd RB/WV A behavioural responses to disturbance. Evidence of a behavioural response to disturbance (e.g. Pease et al., 2005) A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Common Teal Anas crecca Corhwyaden RB/WV A behavioural responses to disturbance. Evidence of a behavioural response to disturbance (e.g. Pease et al., 2005) A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Anas Mallard Hwyaden Wyllt RB/WV A behavioural responses to disturbance. One study shows that distribution of this species is affected by platyrhynchos presence of anglers (Cryer et al., 1987) A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Hwyaden Pintail Anas acuta WV/CB A 1 behavioural responses to disturbance. Evidence of a behavioural response to disturbance (e.g. Pease et Lostfain al., 2005) Hwyaden Wetland/open water habitats relatively inaccessible. Relatively dispersed, although only associated with Garganey Anas querquedula MB/PV A 1 Addfain undisturbed sites. Hwyaden A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review), little direct Shoveler Anas clypeata RB/WV A Lydanbig work on this species however. A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Hwyaden Pochard Aythya ferina RB/WV A behavioural responses to disturbance. There are a range of studies showing disturbance impacts (Cryer et Bengoch al., 1987, Fox et al., 1994, Marsden, 2000, Mori et al., 2001, O'Connell et al., 2007) Hwyaden A variety of studies show disturbance impacts to wildfowl (see Kirby et al., 2004 for review) and show Tufted Duck Aythya fuligula RB/WV A Gopog behavioural responses to disturbance. For studies specifically addressing this species see (Marsden, 2000,

134

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A O'Connell et al., 2007) Hwyaden Occurs in coastal and inland sites but relatively scarce. Impacts of disturbance probably similar to other Scaup Aythya marila WV/CB R Benddu diving ducks. Somateria Hwyaden Evidence that breeding success and behaviour affected by disturbance (Bolduc and Guillemette, 2003, Kay Common Eider RB/RN A mollissima Fwythblu and Gilchrist, 1998, Keller, 1991) Môr-hwyaden Evidence that lower numbers occur off North Wales in areas with higher levels of anthropogenic Common Scoter Melanitta nigra WV/RN R Ddu disturbance (commercial shipping (Kaiser et al., 2006) Môr-hwyaden y Evidence of disturbance impacts on breeding success (Mikola et al., 1994), but unlikely to be any issues Velvet Scoter Melanitta fusca WV A Gogledd with this species during the winter (when present in Welsh waters). Bucephala Hwyaden Goldeneye WV A Disturbance impacts probably similar to other wildfowl (see Kirby et al., 2004 for review). clangula Lygad-aur This species is uncommon. Disturbance impacts probably similar to other wildfowl (see Kirby et al., 2004 Smew Mergellus albellus Lleian Wen WV A for review). Little evidence for disturbance impacts with various studies showing no evidence of disturbance on Red Grouse Lagopus scoticus Grugiar RB A density, breeding success or number shot (Watson, 1979, Lawton, 1990). Requires disturbance-free lekking areas, no evidence that disturbance affects breeding success but the Black Grouse Tetrao tetrix Grugiar Ddu RB R 1 one experimental work did not include high levels of access (Baines and Richardson, 2007, Richardson and Baines, 2004) Disturbance unlikely to be responsible for decline and unlikely to be an issue for this species. Nesting birds Grey Partridge Perdix perdix Petrisen RB R are however prone to desert if disturbed (Crick et al., 2003). Widely dispersed species with only sporadic/occasional occurrence as a breeding bird. Not known if Quail Coturnix coturnix Sofliar MB A 1 particularly sensitive to disturbance. Red-throated Trochydd Gavia stellata WV/PV A Birds at sea likely to be unaffected by shore based disturbance Diver Gyddfgoch Great Northern Gavia immer Trochydd Mawr WV/PV A Birds at sea likely to be unaffected by shore based disturbance Diver Red-necked Gwyach Podiceps grisegena WV A Birds at sea likely to be unaffected by shore based disturbance Grebe Yddfgoch Gwyach Slavonian Grebe Podiceps auritus WV A Birds at sea likely to be unaffected by shore based disturbance Gorniog Black-necked Gwyach Birds at sea likely to be unaffected by shore based disturbance. Some evidence that disturbance effects Podiceps nigricollis WV/FB A 1 Grebe Yddfddu behaviour when close to shore (Liley et al., 2006b). Aderyn-Drycin y Fulmar Fulmarus glacialis RB/PV A Some evidence that cliff-nesting seabirds are vulnerable (Beale and Monaghan, 2005) Graig Aderyn-Drycin Very large colonies in few locations means potentially vulnerable. Nocturnal at breeding sites so Manx Shearwater Puffinus puffinus RB/PV A Manaw disturbance effects likely to be limited. Hydrobates Nocturnal at breeding sites and breeds on remote boulder slopes / beaches so disturbance effects likely to Storm Petrel Pedryn Drycin MB?/PV A pelagicus be limited. Nesting birds prone to desert if disturbed (Crick et al., 2003). Gannet Morus bassanus Hugan MB? A Large colonies at remote island sites, no impacts of disturbance likely. Phalacrocorax Shag Mulfran Werdd RB A Some evidence that cliff-nesting seabirds are vulnerable (Beale and Monaghan, 2005) aristotelis

135

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Known to be sensitive to disturbance: access adjacent to reedbeds, especially on raised banks, might Bittern Botaurus stellaris Aderyn y Bwn WV/FB R 1 displace breeding and wintering birds. Large body of work relating to colonial nesting waterbirds (Carney and Sydeman, 1999), but little or no Little Egret Egretta garzetta Crёyr Bach RB A direct evidence that this species is vulnerable to disturbance. Expanding in the UK and some colonies are in quite disturbed locations Scarce. Restricted to woodlands. Not believed unusually sensitive to disturbance at nest (Roberts et al., Honey Buzzard Pernis apivorus Bod y Mêl MB A 1 1999). Recently reintroduced and expanding population in England. Birds aggregate at winter roosts. No evidence Red Kite Milvus milvus Barcud Coch RB A 1 of particular sensitivity to disturbance when nesting or roosting. Birds often use traditional, localised roosts and can aggregate (including other raptors e.g. Merlin) and Hen Harrier Circus cyaneus Bod Tinwen RB/WV R 1 require disturbance-free areas. Breeding sites likely to be sensitive (Asken Ltd, 2008) . Evidence that disturbance influences both choice of nest site and breeding success (Van Der Zande and Common Kestrel Falco tinnunculus Cudyll Coch RB A Verstrael, 1985) Birds often use traditional, localised roosts and can aggregate (including other raptors e.g. Hen Harrier)and Merlin Falco columbarius Cudyll Bach RB/WV/PV A 1 require disturbance-free areas Largely breeding in coastal areas with evidence of decline in some areas, probably due, in part at least, to Haematopus disturbance/nest trampling. Lots of studies (Coleman et al., 2003, Fitzpatrick and Bouchez, 1998, Goss- Oystercatcher Pioden y Môr RB/WV A 1 ostralegus Custard and Verboven, 1993, Goss-Custard et al., 2006, Stillman et al., 2001, Stillman and Goss-Custard, 2002a, Urfi et al., 1996, Verboven et al., 2001, Verhulst et al., 2001) Recurvirostra Avocet Cambig MB A 1 Nests colonially, largely on saline and brackish lagoons. avosetta Restricted to beaches and clear population consequences of recreational disturbance for this species and Charadrius Ringed Plover Cwtiad Torchog RB/WV/PV A 1 other beach nesting plovers (Lafferty et al., 2006, Liley, 1999, Liley and Sutherland, 2007, Ruhlen et al., hiaticula 2003, Schulz and Stock, 1993, Strauss, 1990, Tratalos et al., 2005, Weston and Elgar, 2005). Distribution during breeding season related to access levels. Lots of studies (Finney, 2004, Finney et al., Golden Plover Pluvialis apricaria Cwtiad Aur RB/WV/PV A 2005a, Pearce-Higgins et al., 2007, Yalden et al., 1989, Yalden and Yalden, 1990) There are many general papers on waders and disturbance and strong evidence of behavioural responses but relatively little work directly on this species (Burton et al., 2002b, Davidson et al., 1993, Fitzpatrick and Pluvialis Grey Plover Cwtiad Llwyd WV/PV A Bouchez, 1998, Hirons and Thomas, 1993, Kirby et al., 1993, Smit and Visser, 1993, Ravenscroft et al., squatarola 2008). This species is territorial in winter (Turpie, 1995) and this means disturbance can have particular consequences. Relatively little evidence that disturbance is an issue during the breeding season (Fletcher et al., 2005) or Lapwing Vanellus vanellus Cornchwiglen RB/WV/PV R the winter (Milsom et al., 1998) Tends to concentrate in large roosts and avoids sites with high levels of boat activity nearby (Peters and Knot Calidris canutus Pibydd yr Aber WV/PV A Otis, 2007); Avoids areas close to footpaths (Burton et al., 2002b). There are many general papers on waders and disturbance and strong evidence of behavioural responses but relatively little work directly on this species (Burton et al., 2002b, Davidson et al., 1993, Fitzpatrick and Purple Sandpiper Calidris maritima Pibydd Du WV A Bouchez, 1998, Hirons and Thomas, 1993, Kirby et al., 1993, Smit and Visser, 1993, Ravenscroft et al., 2008). Evidence of disturbance during breeding season and winter (Burton et al., 2002a, Burton et al., 2002b, Dunlin Calidris alpina Pibydd y Mawn RB/WV/PV R Finney, 2004, Pearce-Higgins et al., 2007)

136

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Lymnocryptes Jack Snipe Gïach Fach WV/PV A Doesn’t flush easily and tends to occur in impenetrable habitats. minimus Gallinago Common Snipe Gïach Gyffredin RB/WV/PV A Doesn’t flush easily and tends to occur in habitats with relatively little access. gallinago Limited evidence for any impacts of disturbance on populations / carrying capacity of UK estuaries during Black-tailed Rhostog 2 Limosa limosa RN/PV/WV/CB R 1 the winter (Gill et al., 2001a, West et al., 2007), but some evidence of avoidance of areas close to Godwit Gynffonddu footpaths (Burton et al., 2002b) There are many general papers on waders and disturbance and strong evidence of behavioural responses Rhostog but relatively little work directly on this species (Burton et al., 2002b, Davidson et al., 1993, Fitzpatrick and Bar-tailed Godwit Limosa lapponica WV/PV A Gynffonfrith Bouchez, 1998, Hirons and Thomas, 1993, Kirby et al., 1993, Smit and Visser, 1993, Ravenscroft et al., 2008). Numenius Whimbrel Coegylfinir PV R Passage migrant, disturbance unlikely to be an issue. phaeopus A variety of studies show behavioural effects of disturbance to wintering birds (e.g. Fitzpatrick and Curlew Numenius arquata Gylfinir RB/WV/PV A Bouchez, 1998); Relatively little work on breeding curlew and disturbance but evidence for lower densities or avoidance of disturbed areas (see Van der Zande, 1984, Haworth and Thompson, 1990b) Pibydd Spotted Tringa erythropus Coesgoch PV/WV A Relatively scare passage migrant and winter visitor, disturbance unlikely to be an issue. Redshank Mannog Evidence of avoidance of areas of suitable habitat close to sources of disturbance (Burton et al., 2002a, Common Pibydd Tringa totanus RB/WV/PV A Burton et al., 2002b) and also some evidence for use of nocturnal use of some sites where disturbance Redshank Coesgoch higher (Burton and Armitage, 2005). Occurs in small numbers and widely distributed across freshwater wetlands. Unlikely to be disturbance Green Sandpiper Tringa ochropus Pibydd Gwyrdd PV/WV A issues due to inaccessibility of the habitats. Common Actitis hypoleucos Pibydd y Dorlan MB/PV A Reduced breeding densities where high access levels (Yalden, 1992) Sandpiper Widely distributed and common. Behavioural responses to disturbance shown to be related to prey Turnstone Arenaria interpres Cwtiad y Traeth RN/WV/PV A abundance (Beale and Monaghan, 2004a) Mediterranean Larus Gwylan Mor y RN/PV/WV A 1 Disturbance only likely to be an issue for breeding gulls, and therefore not for this species Gull melanocephalus Canoldir Gwylan y Common Gull Larus canus PV/WV A Disturbance only likely to be an issue for breeding gulls, and therefore not for this species Gweunydd Lesser Black- Gwylan Colonial breeder. Colonies on accessible coastal sites potentially vulnerable – body of evidence to show Larus fuscus RB/MB A backed Gull Gefnddu Leiaf impacts of access to breeding gull colonies (Hunt, 1972, Robert and Ralph, 1975) Gwylan y Colonial breeder. Colonies on accessible coastal sites potentially vulnerable – body of evidence to show Herring Gull Larus argentatus RB/WV R 1 Penwaig impacts of access to breeding gull colonies (Hunt, 1972, Robert and Ralph, 1975) Great Black- Gwylan Colonial breeder. Colonies on accessible coastal sites potentially vulnerable – body of evidence to show Larus marinus RB A 1 backed Gull Gefnddu Fwyaf impacts of access to breeding gull colonies (Hunt, 1972, Robert and Ralph, 1975) Gwylan Evidence that access at colonies can impact breeding success (Beale and Monaghan, 2004b, Beale and Kittiwake Rissa tridactyla RB/PV A Goesddu Monaghan, 2005, Beale, 2007, Sandvik H and RT, 2001) Sterna Morwennol Limited evidence of disturbance impacts for this species (Dunn, 1975). Colonies likely to be well protected Sandwich Tern MB/PV A 1 sandvicensis Bigddu and access limited.

137

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Morwennol Colonial breeder – largely restricted to Anglesey (Ynys Feurig) and Cemlyn Bay. Disturbance at these sites Roseate Tern Sterna dougallii MB/PV R 1 Wridog managed through wardening etc (see SSSI site management statement). Morwennol Disturbance an issue at sites where access is possible. Some evidence of disturbance impacts probably Common Tern Sterna hirundo MB/PV A 1 Gyffredin also applicable to other tern species (Burger, 1998, Burger, 2003) Colonial breeder – usually associated with beaches and known to be adversely affected by recreational Morwennol Little Tern Sterna albifrons MB/PV A 1 access disturbance, nest trampling and predation (Calado, 1996, Catry et al., 2004, Gochfeld, 1983, Fechan Medeirosa et al., 2007) Guillemot Uria aalge Gwylog RB/WV A Evidence of disturbance impacts on breeding success (Beale and Monaghan, 2005, Thayer et al., 1999) Unlikely to be affected in same way as the above species as tends to nest lower down cliffs and not in such Razorbill Alca torda Llurs RB/WV A tight colonies. Boat traffic may cause impacts?? (See below) Doesn’t occur in tight colonies but some evidence of flushing caused by boat traffic (Ronconi and St. Clair, Black Guillemot Cepphus grylle Gwylog Ddu RB A 2002) Evidence for impacts of disturbance on breeding success for this species and congener; impacts seem to Puffin Fratercula arctica Pal MB A vary between colonies / sites (Pierce and Simons, 1986, Rodway et al., 1996). Nesting birds prone to desert if disturbed (Crick et al., 2003).. Stock Dove Columba oenas Colomen Wyllt RB A Unlikely to be any disturbance effects. No studies known to authors. Nesting birds are particularly prone to desert if disturbed when eggs are fresh (Crick et al., 2003); one Turtle Dove Streptopelia turtur Turtur MB/PV R study shows lower densities in areas with higher recreational use (Van der Zande et al., 1984) Cuckoo Cuculus canorus Cog MB R Unlikely to be any disturbance effects. No studies known to authors. Barn Owl Tyto alba Tylluan Wen RB A 1 No evidence of particular sensitivity to disturbance. Widely dispersed species. Ground-nesting so potentially vulnerable to disturbance, e.g. from dogs. Little evidence for disturbance Short-eared Owl Asio flammeus Tylluan Glustiog RB/WV A effects however (see e.g. Haworth and Thompson, 1990b) Caprimulgus Associated with relatively restricted habitats and known to be adversely affected by recreational access. Nightjar Troellwr MB R 1 europaeus Relatively dispersed across suitable habitat. Common Swift Apus apus Gwennol Ddu MB/PV A Nests in buildings and aerial feeder, unlikely to be any disturbance effects Common Alcedo atthis Glas y Dorlan RB A 1 No evidence of particular sensitivity to disturbance. Widely dispersed species. Kingfisher Green Unlikely to be disturbance effects though as feeds on ground possibly more likely than for other Picus viridis Cnocell Werdd RB A Woodpecker woodpecker species (see below). Disturbance unlikely to be an issue given ecology and habitat use of this species. No direct studies of Lesser Spotted Dendrocopus Cnocell Fraith RB R disturbance and this species, but work on similar Great Spotted Woodpecker has found no correlations Woodpecker minor Leiaf with levels of disturbance and abundance (Ficetola et al., 2007). Some evidence for disturbance effects (Erdos et al., 2009) but a ground nesting species and likely to be Skylark Alauda arvensis Ehedydd RB/WV R affected where access occurs on suitable breeding habitat. There is strong evidence for impacts of disturbance on relatively similar woodlark (Mallord, 2005, Mallord et al., 2007). Gwennol y Sand Martin Riparia riparia MB/PV A Unlikely to be any disturbance effects. No studies known to authors. Glennydd Swallow Hirundo rustica Gwennol MB/PV A Unlikely to be any disturbance effects. No studies known to authors. House Martin Delichon urbica Gwennol y MB/PV A Unlikely to be any disturbance effects. No studies known to authors.

138

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Bondo Corhedydd y Tree Pipit Anthus trivialis MB/PV R Ground nesting species, so potential for impacts of recreation, but no published studies. Coed Corhedydd y Meadow Pipit Anthus pratensis RB/PV A Ground nesting species, so potential for impacts of recreation, but no published studies. Waun Corhedydd y Water Pipit Anthus spinoletta WV A Winter visitor to wetland habitats; unlikely to be any impacts Dŵr Yellow Wagtail Motacilla flava Siglen Felen MB/PV R Unlikely to be any disturbance effects. No studies known to authors. Grey Wagtail Motacilla cinerea Siglen Lwyd RB/PV A Unlikely to be any disturbance effects. No studies known to authors. Common garden bird associated with scrub; disturbance unlikely to be an issue, for example no effect of Dunnock Prunella modularis Llwyd y Gwrych RB A disturbance on density (Van der Zande et al., 1984) Common Phoenicurus Tingoch MB/PV A Mainly breeds in woodland; unlikely to be disturbance effects? Redstart phoenicurus Whinchat Saxicola rubetra Crec yr Eithin MB/PV A No studies known to authors. Common Oenanthe Tinwen y Garn MB/PV A Some evidence of disturbance(Lukac and Hrsak, 2005) but unlikely to be a particular issue for the species? Wheatear oenanthe Declines unlikely to be linked to disturbance (Burfield and Brooke, 2005), and most locations where Mwyalchen y Ring Ouzel Turdus torquatus MB/PV R occurs likely to have relatively low levels of access. “Probably unaffected by disturbance” (Haworth and Mynydd Thompson, 1990b) Fieldfare Turdus pilaris Socan Eira WV/PV R One study shows lower densities in areas with higher recreational use (Van der Zande et al., 1984). No Song Thrush Turdus philomelos Bronfraith RB/WV/PV R evidence of effects of disturbance on nest success (Mayer-Gross et al., 1997) Redwing Turdus iliacus Coch Dan-aden WV/PV R Winter visitor to a range of habitats. Disturbance effects unlikely? Mistle Thrush Turdus viscivorus Brych y Coed RB A No studies known to author. Grasshopper Locustella naevia Troellwr Bach MB/PV R Unlikely to be any disturbance effects. No studies known to authors. Warbler Disturbance unlikely to be an issue for this species. No evidence of effects of disturbance on nest success Whitethroat Sylvia communis Llwydfron MB/PV A (Mayer-Gross et al., 1997) Associated with relatively restricted habitats and known to be adversely affected by disturbance (Murison Dartford Warbler Sylvia undata Telor Dartford RB A 1 et al., 2007, Murison, 2007). Relatively dispersed across suitable habitat. Phylloscopus Wood Warbler Telor y Coed MB R Unlikely to be any disturbance effects? No studies known to authors. sibilatrix Phylloscopus Willow Warbler Telor yr Helyg MB/PV A One study shows lower densities in areas with higher recreational use (Van der Zande et al., 1984) trochilus Regulus Dryw Firecrest RB/PV/WV A 1 Scarce but relatively widely dispersed and perhaps not particularly sensitive to disturbance. ignicapillus Penfflamgoch Spotted Gwybedog Muscicapa striata MB/PV R Unlikely to be any disturbance effects. No studies known to authors. Flycatcher Mannog

139

Natural Species Scientific name Welsh name Status BOCC England: Notes Category A Unlikely to be any disturbance effects. No studies known to authors. Pied Flycatcher Ficedula hypoleuca Gwybedog Brith MB/PV A Unlikely to be any disturbance effects. No studies known to authors. Marsh Tit Parus palustris Titw'r Wern RB R Unlikely to be any disturbance effects. No studies known to authors. Willow Tit Parus montanus Titw'r Helyg RB R

Some strong evidence of disturbance impacts from other European countries, but weak evidence from Pyrrhocorax Chough Brân Goesgoch RB A 1 Welsh study (Blanco et al., 1997, Kerbiriou et al., 2009, Poole, 2003). Keribou et al show disturbance to pyrrhocorax foraging birds (particularly first year birds in late summer) has potential to effect population size. Starling Sturnus vulgaris Drudwen RB/WV/PV R Unlikely to be any disturbance effects. No studies known to authors. Unlikely to be any disturbance effects, one study does show higher densities and higher intake rates at House Sparrow Passer domesticus Aderyn y Tô RB R intermediate levels of disturbance and reduced densities and intake at high disturbance levels (Fernandez- Juricic et al., 2003) Golfan y Tree Sparrow Passer montanus RB R Nesting birds are particularly prone to desert if disturbed when eggs are fresh (Crick et al., 2003) Mynydd Carduelis Disturbance unlikely to be an issue for this species. No convincing evidence of effects of disturbance on Linnet Llinos RB R cannabina nest success (Mayer-Gross et al., 1997) Carduelis Twite Llinos y Mynydd RB/WV R “Probably unaffected by disturbance” (Haworth and Thompson, 1990b) flavirostris Llinos Bengoch Lesser Redpoll Carduelis cabaret RB R Unlikely to be any disturbance effects. No studies known to authors. Lieaf Disturbance unlikely to be an issue for this species. No convincing evidence of effects of disturbance on Bullfinch Pyrrhula pyrrhula Coch y Berllan RB A nest success (Mayer-Gross et al., 1997) Coccothraustes Hawfinch Gylfinbraff RB/PV R Nesting birds are particularly prone to desert if disturbed when eggs are fresh (Crick et al., 2003) coccothraustes Plectrophenax Snow Bunting Bras yr Eira PV/WV A Unlikely to be any disturbance effects. No studies known to authors. nivalis Yellowhammer Emberiza citrinella Melyn yr Eithin RB R Nesting birds are particularly prone to desert if disturbed when eggs are fresh (Crick et al., 2003). Emberiza Reed Bunting Bras y Cyrs RB A Unlikely to be any disturbance effects. No studies known to authors. schoeniclus Corn Bunting Miliaria calandra Bras yr Ŷd RB/PV R Unlikely to be any disturbance effects. No studies known to authors.

140

Appendix II: Simplifying the Phase I to give habitat layers for the vulnerability map In the literature review we used simple broad habitat categories. The Phase I habitat map for Wales represents the best habitat dataset and therefore in this table we cross reference the Phase I habitat types to the simple broad categories used in the review. The table below sets out how we propose to group Phase I habitats according to the . .

Further split / Broad Habitat type used Code Habitat refinements necessary in review to map Woodland and scrub A.1.1.1 semi-natural broadleaved woodland A.1.1.2 planted broadleaved woodland A.1.3.1 semi-natural mixed woodland 1 Broadleaved, mixed and A.1.3.2 planted mixed woodland yew woodland A.4.1 felled broadleaved woodland A.4.3 felled mixed woodland A.1.2.1 semi-natural coniferous woodland A.1.2.2 planted coniferous woodland 2 Coniferous woodland A.4.2 felled coniferous woodland A.2.1 dense scrub A.2.2 scattered scrub 1 Broadleaved, mixed and A.3.1 scattered broadleaved trees yew woodland A.3.2 scattered coniferous trees A.3.3 scattered mixed trees Grassland and marsh B.1 acid grassland Upland grassland would be B.1.1 unimproved acid grassland 8 Acid grassland split simply as improved or unimproved B.1.2 semi-improved acid grassland B.2.1 unimproved neutral grassland Upland grassland would be 6 Neutral grassland split simply as improved or B.2.2 semi-improved neutral grassland unimproved B.5 marshy grassland B.5.1 marshy grassland Juncus dominated 11 Fen, marsh and swamp B.5.2 marshy grassland Molinia dominated B.3.1 unimproved calcareous grassland Upland grassland would be 7 Calcareous grassland split simply as improved or B.3.2 semi-improved calcareous grassland unimproved B.4 improved grassland 5 Improved grassland Upland / lowland split Tall herb and fern C.1.1 bracken 9 Bracken C.1.2 scattered bracken Within surrounding habitat C.2 upland species rich ledges 16 Inland rock 17 Built-up areas and C.3.1 tall ruderal herb gardens C.3.2 non-ruderal herb and fern 16 Inland rock Heathland D.1.1 dry acid heath D.1.3 scattered dry heath 10 Dwarf shrub heath Split into lowland or upland D.2 wet heath D.3 lichen/bryophyte heath

141

Further split / Broad Habitat type used Code Habitat refinements necessary in review to map D.5 dry heath/acid grassland mosaic D.6 wet heath/acid grassland mosaic D.1.2 dry basic heath basic dry heath/calcareous grassland D.7 mosaic Mire E.1.6.1 blanket bog E.1.6.2 raised bog 12 Bogs E.1.7 wet modified bog E.1.8 dry modified bog E.3.1.1 modified valley mire 11 Fen, marsh and swamp E.3.2 basin mire E.3.2.1 modified basin mire E.4 bare peat 12 Bogs E.3.1 valley mire E.2 flush and spring E.2.1 acid/neutral flush E.2.2 basic flush E.2.3 brophyte-dominated spring 11 Fen, marsh and swamp E.3 fen E.3.3 flood-plain mire E.3.3.1 modified flood plain mire Swamp, marginal and inundation F.1 swamp 11 Fen, marsh and swamp F.1.1 scattered swamp Within surrounding habitat 13 Standing open water F.2.2 inundation vegetation and canals Open water 13 Standing open water G.1 standing water and canals G.2 running water 14 Rivers and streams Coastland 6 Mudflats and sandy H.1.1 intertidal mud/sand beaches H.1.2 intertidal cobbles/shingle 2 Coastal vegetated shingle H.1.3 intertidal rocks/boulders 3 Rocky shores H.4 rocks/boulders above mhw H.2.4 scattered salt marsh plants 5 Saltmarsh H.2.6 salt marsh 6 Mudflats and sandy H.3.1 mud/sand above mhw beaches H.3.2 shingle/gravel above mhw 2 Coastal vegetated shingle H.6.4 dune slack H.6.5 dune grassland H.6.6 dune heath 7 Coastal sand dunes H.6.7 dune scrub H.6.8 open dune H.8.1 hard cliff 16 Inland rock H.8.2 soft cliff 4 Maritime cliffs and slopes H.8.4 coastal grassland H.8.5 coastal heath 10 Dwarf shrub heath

142

Further split / Broad Habitat type used Code Habitat refinements necessary in review to map coastal heath/coastal grassland H.8.6 mosaic Rock exposure and waste I.1 natural rock exposure I.1.1 inland cliff I.1.1.1 acid/neutral inland cliff I.1.1.2 basic inland cliff I.1.2 scree I.1.2.1 acid/neutral scree 16 Inland rock I.1.2.2 basic scree I.1.3 limestone pavement I.1.4 other rock exposure I.1.4.1 acid/neutral rock I.1.4.2 basic rock I.1.5 cave I.2.1 quarry not allocated Separate category for I.2.2 spoil not allocated industrial habitats? I.2.3 mine not allocated I.2.4 refuse-tip not allocated Miscellaneous J.1.1 arable 4 Arable and horticulture J.1.2 amenity grassland 5 Improved grassland J.1.3 ephemeral/short perennial not allocated J.1.4 introduced scrub not allocated J.1.5 gardens 17 Built-up areas and J.3.4 caravan site gardens J.3.6 buildings J.3.5 sea-wall not allocated J.3.7 track (not comprehensively digitised) not allocated J.4 bare ground not allocated NA not accessed land not allocated habitat code illegible on the original ? not allocated vegetation map

143

Appendix III: Example maps The following four maps are all at the same scale and show some simple, indicative examples of how the vulnerability map might be derived. Map 1: Natmap soil data (vector data) for Wales Map 2: Example of Phase I habitats, for clarity just four habitats are chosed. Map 3: Example of habitat scoring, 500m grid cells with scores for each cell derived from the four habitats in map two and a hypothetical score given for each. Map 4: Example of habitat scoring combined with species data.

144

Map 1 Map of soil types within Wales, taken from the National Soil Map (natmap vector)

145

Map 2 All habitats in Wales classed as broadleaved, mixed and yew woodland (brown), coniferous woodland (dark green), acid grassland (light green) and lowland heathland (purple) in the Phase 1 habitat survey. Lowland heathland is mapped using the 300m contour in this example.

146

Map 3 A grid has been constructed over the selected habitats from Map 3 and a habitat score derived from the type of habitat within each cell (Broadleaved woodland = 3, acid grassland = 3, coniferous woodland = 1 and heathland = 5) and the proportion of the cell covered by that habitat. Colours are flexible and have been graded to match those in Map 4. As a result the key indicates scores which would not be achievable by habitat score alone but are comparable with Map 4.

147

Map 4 Total score for each grid cell, similar to that in Map 3, but also incorporates species scores.

148

149