Grouse Moor Management Group: report

4. Summaries of scientific evidence on raptor persecution and predation, muirburn, Mountain Hares, and use of medicated grit

In this section we summarise the scientific evidence on raptor persecution and predation, the practice of muirburn, the ecology and management of Mountain Hares and the use of medicated grit. We have gathered evidence from the following sources: monographs, peer-reviewed journal articles, reports from relevant public sector organisations and NGOs, responses to our questionnaire, information supplied by Police Scotland and the Crown Office and Procurator Fiscal Service and oral evidence from presentations and invited experts (see Annex 2 for further details). As in previous sections in we do not generally cite sources in the main body of the report, but list all the published sources in Annex 1.

Raptor persecution and predation

Background

This review was triggered, in part, by the SNH report Analyses of the fates of satellite tracked Golden Eagles in Scotland which concluded that “a relatively large number of the satellite tagged golden eagles were probably killed, mostly on or near some grouse moors where there is recent, independent evidence of illegal persecution” (Whitfield & Fielding, 2017, p. vii). Prior to this, extensive research had explored the relationship between raptors and predation on grouse – most notably the initial Joint Raptor Study (JRS) at Langholm and the more recent Langholm Moor Demonstration Project (LMDP). Scotland’s Moorland Forum report Understanding Predation (2016) provides a wider review of the role that predators play in terms of overall ecosystem health. The scientific evidence on the persecution of raptors and the relationship between raptors and predation on grouse is summarised in this section alongside comments on gaps and uncertainties in the evidence-base.

Effects of illegal killing on raptors

Owing to persistent but then legal killing of many raptor species, mainly in the interests of game rearing, raptor numbers across Britain were greatly reduced in the 19th-early 20th centuries. Five species were eliminated altogether from Britain, and most others were much restricted in distribution. All species became protected by law from 1954, except for the Sparrowhawk which was protected from 1968 in Scotland. Around this time, use of organo-chlorine pesticides caused further reductions in the numbers of several species. Recoveries in numbers followed reductions in organo-chlorine use and deliberate killing, and have been especially evident since the 1970s. At the national level most species have increased substantially over recent decades, but most still do not occupy their entire potential range in Britain. Eliminated species either recolonised naturally or were reintroduced. Despite increasing nationally, several raptor species have declined in grouse moor areas during this century.

Raptors vary in the extent to which they eat the eggs, chicks or adults of the International Union for the Conservation of Nature’s (IUCN) red and amber-listed ground-nesting birds, including Red Grouse. Important grouse-predators include the Golden Eagle, Peregrine Falcon, Goshawk and Hen Harrier. Three of these species are green-listed, but the Hen Harrier is red-listed. Studies have shown that illegal killing is reducing the population and breeding success of all these species (plus Red Kite) in at least some grouse moor areas. The evidence consists of: (1) observed population declines or reduced occupancy of known territories; (2) reduced nest success; (3) reduced adult survival; (4) reduced age of first breeding (implying ready availability of vacant territories); (5) unexpectedly high disappearance of satellite-tracked raptors on grouse moors; and (6) finding on moors of poisoned baits and traps, and shot or poisoned carcasses of raptors. Not all types of evidence are available for every species. As well as being targeted directly, some raptors are killed incidentally in attempts to trap or poison other predators. Killing on driven grouse moors can be inferred to be affecting raptor numbers over much wider areas. In the absence of interference, all these raptors breed as well or better on grouse moors than in other upland habitats, benefiting from various aspects of moorland management. Discovered cases of raptor killing probably represent only a small proportion of actual cases. Then only a small proportion of discovered cases get to court, and an even smaller proportion lead to successful convictions.

During this century, reports of poisoned or shot raptors have declined in numbers. However, this cannot be taken as evidence of a decline in raptor killing. It may be associated with a reported (but not proven) increase in the use of night-viewing equipment to shoot raptors on their roosting sites. Night-killing can be practised at any time of year, and carcasses removed, making legally robust evidence of illegal killing practically impossible to obtain. Likewise, the repeated disappearance of satellite-tracked birds on particular moors is not sufficient to obtain a prosecution in a criminal court, because the disappearance cannot be assigned to a particular individual, and no carcass is normally available. These changes over the last two decades have made it increasingly difficult to prosecute anyone for killing raptors. However, a continued decline in reported cases has followed the introduction of vicarious liability in Scotland: the UK-wide Birdcrime report shows five confirmed incidents of illegal raptor killing in Scotland in 2017 compared with 27 as the last five year average (but the figure had increased to 12 in 2018). The 2017 report also drew attention to the relatively large number of unexplained disappearances of satellite-tracked raptors in grouse moor areas. It is also possible (but unproven) that perpetrators have become more concerned with hiding the evidence of their actions in recent years, following press reports of the disappearance of satellite-tracked birds. Some moor managers may have ceased or reduced their killing of raptors in recent years, although other evidence suggests that the practice continues.

Effects of raptors on grouse

The major predators on grouse (eggs, chicks or adults), namely Fox, Stoat, Weasel and Carrion/Hooded Crow, are routinely and legally killed on grouse moors. This leaves birds of prey as the principal remaining predators, so it is not surprising that most discovered grouse remains show signs of having been killed (or at least fed upon) by raptors. In this situation, with Foxes, Stoats, Weasels and Crows controlled, scientifically robust evidence is available from the first Langholm study (JRS) to show that, in sufficient numbers, Hen Harriers can reduce the densities of grouse to such low levels that driven grouse shooting is impracticable. The same is likely to be true for Peregrines in some areas, but in the only relevant study, Peregrines were preying on the same grouse population as harriers. No detailed studies have been made of the impact of Golden Eagles and Goshawks on Red Grouse. One 6-year study at Langholm suggests that individual Buzzards have negligible impact on grouse, although at high densities promoted by abundant alternative food supplies, their collective impact could become significant.

Langholm Moor Demonstration Project

The second study at Langholm (LMDP) was a ten-year project which sought to recover a former driven grouse moor to commercial viability. The project has now ended and its final report Managing Moorland for Birds of Prey and Red Grouse (Langholm Moor Demonstration Project Board, 2019) has been published together with several scientific papers. When coupled with the Langham Moor Demonstration Project: seven year review (Langham Moor Demonstration Project, 2014) this report provides an authoritative guide on the challenges of reconciling commercial grouse shooting with healthy raptor populations

The main conclusions were:

• The grouse showed good health, body condition and large clutch size;
• Habitat was not a limiting factor for any major aspect of grouse performance;
• All legal predator control was carried out to good industry standards;
• In some years, with high vole numbers, around 70 nests of raptors, owls and Ravens were found on or near the 115 km2 grouse moor;
• Under the control of mammalian predators, 82% of grouse kills found were assigned to raptor predation or scavenging, but it was not possible to assign kills to individual species;
• Grouse did not achieve sufficient numbers to be shot on a commercial driven basis and the keepering team was made redundant before the end of the project;
• Other red and amber-listed ground-nesting species such as Lapwing, Curlew and Golden Plover increased slightly in numbers during the study, but did not reach the conservation targets set; Meadow Pipits and Black Grouse increased more substantially.

Gaps and uncertainties in the evidence-base

The first study at Langholm Moor (JRS) showed that, in the absence of killing, Hen Harriers increased to levels at which their predation prevented grouse from reaching numbers sufficient for driven shooting. For various reasons Langholm moor provided ideal habitat for harriers. But it is uncertain, in the absence of killing, how many other moors would support harriers at densities high enough to cause similar suppression of grouse numbers. It is partly fear over a repeat of the Langholm experience that leads other moor managers to kill harriers and other raptors.

The years between the JRS and the LMDP, when gamekeepers were absent, showed that the resulting lack of habitat management and predator control was associated with reduced numbers and breeding success of ground-nesting harriers and other species. It has not yet been quantified to what extent similar cessation of grouse moor management elsewhere has negatively affected ground-nesting raptors, but their downward population trends in a number of areas which were formerly grouse moors suggests similar effects may be operating.

The Langholm experience has done much to shape the attitudes of land managers and, as noted above, the final report of the LMDP is an authoritative guide on the relationships between commercial grouse shooting and healthy raptor populations. But, as in all field experiments, the question arises as to how far Langholm Moor is typical of Scottish grouse moors in general.

Studies have so far concentrated on the main predators of grouse. There remains the possibility that, while no individual species would have a significant impact on grouse, the collective impact of several or all species might together reduce grouse to levels below those needed for driven shooting. This is a field of study requiring enormous resources, and impinges on the question of to what extent removal of one key predator species might lead to greater predation by others. It is also unknown what impact the increase in the national population of Buzzards and Ravens is having on available food sources and if this is impacting on other raptors who may have to feed more on grouse and other ground-nesting birds than they otherwise would.

The JRS study implied that Peregrines could also have a substantial impact on grouse numbers, but it was impossible to separate their winter impact from that of harriers. The individual impact of Peregrines in areas lacking harriers therefore remains unknown. In similar vein, the impact of raptors studied in the LMDP was made in a situation of no grouse shooting.

It has been claimed that Golden Eagles, by killing or deterring other raptors, can limit the densities of these other species in their home range. This situation needs more study, because if true, it provides a natural way in which the numbers of some raptors could be controlled, and the overall predation on grouse reduced.

Diversionary feeding

This practice, adopted as part of the LMDP, involves the daily provision of extra food (in the form of small dead animals) to raptor pairs to see whether it would lessen their predation on grouse to an acceptable level. Diversionary feeding of Hen Harriers while they were raising young resulted in a substantial reduction of their predation on grouse both during an initial trial and during the LMDP. But there are concerns from the shooting community that the practice is time consuming and could result in additional harriers on the moor in subsequent years. Also, the number of grouse did not increase sufficiently to allow shooting, at least not under the stated threshold densities of grouse considered necessary by the project managers for driven shooting. From a land manager’s perspective, the known and potential disadvantages of diversionary feeding out-weigh the advantages, so this is not seen as an effective management tool, although others disagree.

Brood management

As part of the Department of Environment Food & Rural Affairs’ Hen Harrier Recovery Plan, a 2-year trial granted under a research licence began in England in 2019 aimed at increasing the overall population of Hen Harriers. Any moor manager who has more than a specified density of Hen Harrier nests on their land is allowed to arrange for the surplus clutches or broods to be collected for captive rearing in a dedicated facility, and the release of the resulting young elsewhere. This partly removes the concern of the moor manager that, if he allows harriers to nest on his land, in time their numbers on the moor might increase to such a level as to make driven shooting non-viable. The removal of broods means that no grouse are killed to feed them. In this opening year, only one brood became available for hand rearing and, although five young fledged successfully, it is too early to judge whether such a trial can achieve its objective of increasing the numbers of harriers nesting on English moors. The scheme could fail if moor managers refused to participate in the scheme, or if too many reared birds were killed before they could breed themselves.

Muirburn

Background

Most moorland vegetation is highly flammable and ‘muirburn’ refers to the burning of vegetation in moorland areas, usually in a controlled manner, in order to maintain open moorland. This practice has been a favoured management tool for many centuries. The iconic landscape in which heather is the dominant species, much appreciated by tourists during the summer months and designated for its international conservation importance, is partly natural (beyond the climatic limits for tree and scrub growth) and partly a ‘cultural’ landscape kept open by active management.

There is a particularly strong association between muirburn and habitat management for Red Grouse, but muirburn is also used in some areas for the management of deer and livestock grazing. On moorland areas managed for driven grouse shooting, rotational muirburn is carried out to create small patches of heather of different ages to produce patches of ground containing young, more nutritious heather shoots for grouse to eat and patches of taller heather for cover – the aim being to produce a mosaic in which heather of different heights/ages occurs within the territory of each grouse pair. Well-managed muirburn normally achieves its desired aims of providing good habitat for grouse and other species. But the wider impacts of muirburn are highly contested, with variable and sometimes contradictory findings from different experiments and monitoring work. Guidance on the management of muirburn is available in the Muirburn Code, administered by SNH on behalf of the Scottish Government. As noted in the Code: “Most of Scotland’s moorland is not burnt or cut and does not require burning, but fire and cutting equipment are useful management tools, when used with skill and understanding. The Scottish Government supports well-managed muirburn and recognises its potential to reduce the impact of wildfire” (Scottish Government, 2017, p. 1).

Muirburn Code

The Muirburn Code (most recently updated by Scotland’s Moorland Forum in 2017) is designed to promote best practice and minimise the likelihood of the detrimental impacts that can arise from muirburn. The Code provides detailed guidance, with some statutory provisions relating only to the timing and notification of a proposed muirburn (derived from the Hill Farming Act 1946 as amended by the Wildlife and Natural Environment (Scotland) Act 2011 and the Climate Change (Scotland) Act 2009). In addition to complying with these statutory provisions, landowners receiving payments from the Rural Payment and Inspectorate Directorate Payments (RPID) must also meet requirements in terms of Good Agricultural and Environmental Condition (GAEC), some of which relate to elements of the Muirburn Code (see below). Failure to meet these requirements can result in a reduction being applied to support payments.

Two major issues arise in terms of compliance with the Muirburn Code. First, under RPID cross-compliance, although breaches can be investigated and penalty charges applied, only a small subset of muirburn-related activities in the Code are deemed to ‘breach’ the grant conditions – specifically a fire left unattended, an uncontrolled fire over a large area, or burning which results in damage to woodland. Other potentially damaging activities listed in the Muirburn Code, such as burning on steep slopes, thin soils or peatlands, are currently not included in this penalty system. Given the effort and resources currently being put into peatland restoration, the omission of burning on peatlands from this penalty system is particularly concerning. Second, the ‘breach and penalty’ system only applies to those receiving RPID muirburn-related support payments. No other penalty systems, other than prosecution for failing to notify neighbours or undertaking muirburn outwith the specified time periods, address non-compliance with the Muirburn Code. Both issues represent major weaknesses in promoting the sustainable management of muirburn.

Although the Muirburn Code is designed to minimise the likelihood of detrimental impacts by promoting ‘good’ practice, there have been no assessments to date of levels of adherence to the good practice detailed in the Code. In the absence of a robust system of monitoring compliance, plus the relatively few instances under the Code of the withholding of RPID payments under cross compliance, it is not currently possible to assess the Code’s effectiveness.

Understanding the impacts of muirburn

Reviewing the extensive literature on muirburn impacts immediately reveals the importance of fire characteristics in determining biodiversity and wider ecosystem effects, yet in much of the debate over contradictory findings, this is sometimes glossed over or overlooked. Greater recording of fire characteristics alongside their impacts is much needed. In general terms, the benefits of muirburn in providing young, more nutritious shoots for grouse (and livestock, deer and Mountain Hares) are well-established. There is also evidence that regular muirburn managed in accordance with the Muirburn Code can increase above-ground biodiversity (evidence includes plants, birds, invertebrates) compared with unburnt moorland, particularly in dry heaths, through the creation of mosaics of different ages of heather giving a mix of habitat structures. Muirburn does of course restrict colonisation by woodland that would represent the natural habitat type in many of these ‘cultural’ moorland areas. There is also strong evidence that muirburn can cause detrimental effects in some situations – on biodiversity, hydrology, soil stability and other components of the system. Many factors can determine the impact of a moorland fire, but possibly one of the most critical factors is fire intensity (driven by fuel load, weather, moisture content and many other factors). To date this has rarely been recorded, so in very few cases can direct causal-associations be made between fire intensity and impacts. A key issue is whether or not the fire has burned into the moss/litter layer/soil/peat – in that order it greatly increases the likelihood of detrimental impacts.

Muirburn impacts can also differ according to the type of moorland. The strongest, but still inconclusive evidence for a greater likelihood of long-term detrimental impacts comes from blanket bog/wet heath areas, and it has been widely assumed that regular muirburn is detrimental to peat-forming plant species. However, this is not conclusive as several studies have found the opposite, including a long-term (60 years) experimental study in the Pennines.

The effects of muirburn below-ground are the least well understood. In general terms the intensity of a fire is thought to be a key issue: impacts on soil structure/properties (and carbon in particular) tend to be more severe (sometimes catastrophic) under more intense, longer duration fires igniting dry soil/peat. Fires penetrating the moss/litter soil are likely to destroy much of the seed bank, which could result in a very different post-fire plant species composition as well as changes to hydrology, soil chemistry and increased likelihood of erosion. Muirburn can have both positive and negative effects on carbon storage, both directly, by affecting carbon contents of soil and vegetation, and indirectly, by affecting carbon storage potential through the changes in plant community composition after fire. There is often an assumed net loss of carbon under regular muirburn, but the evidence is not conclusive and the longest running study (60 years) shows reduced carbon sequestration in regularly burnt areas but no long-term reductions in soil carbon accumulation. Data on muirburn impacts on dissolved organic carbon, particulate organic matter, suspended sediments, aluminium, iron etc. in runoff are also inconclusive, with varying data on losses and subsequent recovery within the system as vegetation regrows, as well as some evidence for interactive effects with historical and current pollutant deposition and post-fire rainfall patterns. The few studies on hydrology indicate lowering of water tables in regularly burnt areas, but divergent effects on overland water flows according to the nature of concurrent rainfall events. There are obvious implications here for future climate change affecting muirburn impacts.

Given the clear, but unquantified, risks of major environmental damage from moorland fires, it is appropriate and critically important to increase the assessment of fire practices and their impacts (both positive and negative), particularly in relation to predicted climate changes. We note the recent fire assessment work by SNH and others using remote sensing and are highly supportive of this as a powerful approach to assess location, frequency and extent of moorland fires, combined with more detailed monitoring on the ground, particularly in relation to the ‘best practice’ specified in the Muirburn Code.

Some key questions

Is burning necessary to retain heather-dominance?

It was traditionally thought that regular muirburn was necessary for heather to remain dominant within the ‘cultural’ moorland zone, but this has been disproved at least for some moorland areas where heather has remained dominant (with the plants ‘rejuvenating’ through stem layering/rooting) for at least 40-60 years without burning. It is not known what proportion of Scotland’s moorland has never been burnt so this is a key evidence gap – we cannot estimate what proportion of moorland might remain heather-dominated in the long-term in the complete absence of fire. This information would greatly inform the debate about the ‘need’ or otherwise of regular moorland burning in terms of maintaining heather dominance (as opposed to maintaining a supply of young, more nutritious shoots for grazing, as mentioned above). Clearly, in areas where tree seed sources are sufficiently abundant then moorland areas might gradually become (re)colonised by young trees. But this process would likely be relatively limited in areal extent in the short-to medium-term because it would depend mainly on short distance spread from existing trees.

Burning versus cutting

There is relatively little comparative information on the impacts of heather burning versus cutting and from the few studies found, the differences are apparently not simple. In relation to heather regeneration (the main focus of most published comparisons), old heather stands tend to regenerate better after fire than cutting (strongly dependent on seed germination and seedbed), but the few data available indicate that younger heather may regenerate better under cutting than burning. Discussions on the relative impacts of muirburn versus cutting on other elements of biodiversity are mostly speculative as there is so little information available. It is important also to note that burning and cutting are not interchangeable everywhere – for example machine-cutting is not practicable on slopes that are too steep or too rocky for machinery to operate safely.

Climate change, muirburn and wildfires

The risk of uncontrolled fire is predicted to increase in Scotland as the summers are projected to become hotter and drier under current climate change scenarios. Muirburns can potentially have both negative and positive effects in this respect. They can of course be the cause of wildfires if they get out of control, but they can also be used to reduce the risk of wildfires through reducing fuel-load build-up or acting as fire-breaks. Data sources from both Scotland and England are relatively few and quote varying proportions of wildfires starting from muirburns, and the levels of risk are currently difficult to quantify. This is primarily due to the uncertainty (stated by many including Scottish Fire and Rescue Service, SFRS) regarding attribution of causes of wildfires, coupled with the fact that an unknown number of out-of-control management fires are brought back under control without reporting (a 2003 questionnaire to 42 estate owner/managers reported that less than half the wildfires on their land involved call-outs to the fire brigade). A very recent unpublished assessment by the Scottish Wildfire Forum of ten years of SFRS data found that, out of 118 fires attended by SFRS, less than 10% of reported wildfires were attributed to ‘controlled burning’ or ‘heather burning’ (on any land, not just grouse moors), but larger numbers were reported as ‘other – not known’ cause. The only published data we found that came directly from estate owners/managers was the 2003 questionnaire (as above). Responding estates reported an average of 1-2 wildfires per estate in 2003, less than 50% of which were caused by escaped management fires. Putting this into the context of the large numbers of management fires carried out, this represented less than 1% of all management fires that year on these estates, but the average size of each wildfire was 150 ha, i.e. considerably larger than a normal management fire. Minimising risk is of course paramount and it is critically important that the Muirburn Code and regulation updates relating to muirburn use the best available evidence to minimise the chance of muirburn fires getting out of control. Muirburn and grazing and cutting can all play a role in reducing fuel loads and possibly reducing the incidence or at least severity (temperature) of wildfires, although there is a lack of data demonstrating how these management actions can be combined to best effect. We note the recent initiative announced by the SFRS to explore and test the use of muirburn as a management tool to reduce wildfire risk in future, and the current work funded by the Scottish Government to explore the possible development of a fire danger rating system.

Mountain Hares

Background

Mountain Hares are widespread in Scotland, but they are particularly strongly associated with heather moorland, including areas managed for driven grouse shooting where their populations are sometimes harvested and controlled. Mountain Hares are also considered to benefit from some aspects of grouse moor management, particularly the reduction in predators and the creation/maintenance of mosaics of different ages of heather.

Since the 19^th century, Mountain Hares have been a legal game species for sport shooting during the open season: 1^st August - 28^th February. Shooting of large numbers of hares on some estates has received much attention in the media and this is one of the drivers for including Mountain Hares in our evidence review and recommendations. A licence is required from SNH to kill hares during the closed season and landowners must provide a justification and indication of numbers to be killed. In addition to sport shooting, Mountain Hares are also killed for the protection of young trees and other sensitive plants, and (from our questionnaire responses) they are still killed on some estates as part of tick control measures (see discussion below). The number of Mountain Hares that can be killed in the open season is not regulated and does not require any statutory reporting, but land managers are expected to exercise restraint, in regard of our obligations under the EC Habitats Directive.

Conservation status of Mountain Hares

The Mountain Hare is on the Scottish Biodiversity List, i.e. considered by Scottish Ministers to be of ‘principal importance’ for biodiversity conservation. The Mountain Hare is also listed in Annex V of the EC Habitats Directive as a species “of community interest whose taking in the wild and exploitation may be subject to management measures”. Member States are required to ensure that the exploitation of Annex V species “is compatible with their being maintained at a “favourable conservation status” and to make regular reports on this to the EU.

In a recent assessment of all UK mammals carried out by the Mammal Society according to IUCN red listing criteria, the conservation status of Mountain Hares in Scotland was assessed as “Near Threatened” (but close to “Vulnerable” under several of the criteria). The assessment noted that “further evidence is urgently required since re-evaluation may move the species to the Vulnerable category.” In the Joint Nature Conservation Committee 2019 report to the EU for the period 2013-18, Mountain Hares in the UK were categorised as being in an “unfavourable-inadequate conservation status”; the term ”inadequate” referring to a lack of data (see also Annex 4).

A major issue in determining the conservation status of Mountain Hares is a poor evidence-base. Until recently (see below) there has been no standardised method specifically designed for counting Mountain Hares, and there is also no mandatory formal monitoring of Mountain Hare populations in the UK or Scotland. All published estimates of Mountain Hare numbers to date are at least partially based on ancillary data and primarily non-hare-specific surveys. The recent publications on Mountain Hare numbers illustrate this problem well. The Mountain Hare data recorded as part of BTO’s Breeding Bird Survey is based on daylight counts by observers on foot; the recent papers by Watson & Wilson and Hesford et al. both used data from daylight counts made by observers with dogs. Given that the study locations selected for these papers were non-random, we cannot extrapolate the findings beyond those particular areas and sites. As detailed by Newey et al. (2018) in the SNH report Developing a counting methodology for mountain hares (Lepus timidus) in Scotland, these methods are not particularly well suited to surveying Mountain Hares and have no known calibrations with actual population sizes, nor any information on repeatability-accuracy. Given the problems with methodology used to date, the lack of whole-Scotland count data, and the highly contrasting findings presented by different papers, it is currently not possible to estimate with any certainty what the population of Mountain Hares in Scotland actually is, nor how it has changed over time. This is a critical issue for this Annex V species.

There is also no mandatory formal recording of Mountain Hare numbers shot in-season. The National Gamebag Census administered by the Game & Wildlife Conservation Trust (GWCT) is voluntary and records numbers reported as shot by participating estates but does not record numbers present. Since 1961, we understand that an average of 30% of the total number of estates that have reported shooting hares return Mountain Hare ‘bag’ information in any one year but it is not known what proportion of the others are ‘non-returns’ and what proportion are ‘non-shoot’ years. We have been informed by GWCT that analyses of data from a questionnaire on hare presence / absence and numbers shot that was sent to estates across Scotland in 2016-17 is currently in preparation.

Uncertainties over population size, trends over time and the number of Mountain Hares being killed undermine any attempt at a robust and reliable assessment of the conservation status of Mountain Hares. Accordingly, we support the roll out of a standardised national counting method (coupled with mandatory reporting of numbers killed – see below) and welcome SNH’s new standardised counting method (Newey et al. 2018), currently being rolled out to moors by GWCT, as a first step in resolving the debate on trends in Mountain Hare numbers. At present the understanding of population responses to numbers killed is too poor to impose direct controls on shooting of Mountain Hares with any degree of certainty.

Management of Mountain Hare populations

The sporting sector manages Mountain Hare populations by shooting during the open season for sporting purposes, as well as to reduce numbers when considered by the estate to be necessary. There is controversy around Mountain Hare population management. Below we list the main reasons given for managing Mountain Hare populations (collated from verbal and/or written evidence presented to the Review Group, as well as published literature) and we consider each in turn:

• Sport shooting and game food;
• Reducing competition with grouse for food (heather shoots);
• Reducing browsing impacts on young trees and other sensitive plant species;
• Reducing the tick burden on red grouse and, associated with this, reducing the incidence of Louping Ill virus within the local grouse population;
• Reducing parasite burdens on Mountain Hares and risk of wildlife diseases; both potentially causing population decline.

Impact of sport shooting on hare populations

Having reviewed the literature and taken evidence from key stakeholders, it is widely assumed that the number of hares killed for sport shooting probably has a limited effect on Mountain Hare conservation status overall, although this could differ locally. This finding is based on the assumption that traditional sport shooting cull levels are designed to be proportional to what local populations are believed to be able to sustain, with the intention of sustaining the sporting interest into the future. But we note that this assumption cannot be tested on account of the very limited evidence available.

Assessing the impact of sport shooting is further complicated by Mountain Hares having the potential for high population growth, although this can vary significantly between years. A modelling-based study, parametrised with data from a Scottish grouse moor managed for driven grouse shooting, suggests that Mountain Hare populations may be robust to a maximum of 40% of individuals removed annually by whatever means including natural predation. Against this estimate many other factors need to sit (e.g. population size, age of hares shot, other concurrent causes of mortality, etc.) which, when combined with limited dispersal capacity, means that local extinctions could happen as a result of sport shooting culls.

Reducing food-competition with grouse; browsing impacts on young trees and other sensitive plant species

There is no substantive evidence that Mountain Hare grazing of heather will detrimentally affect grouse through food-competition. Calculations of potential offtake of heather shoots by hares also confirm that this would be extremely unlikely even at maximum hare population densities found in Scotland.

There is evidence that Mountain Hare browsing activity can locally reduce or suppress tree and shrub growth, i.e. preventing natural succession and contributing towards maintaining open heather moorland. SNH reported that they have issued out-of-season licenses to cull Mountain Hares to protect young trees, but never yet to protect any open-ground plant species.

Hare population control as part of tick control measures

There is no substantive evidence to support the population control of Mountain Hares as part of tick and/or Louping Ill virus control to benefit grouse, except under unusual circumstances. The published study that was used to support this assertion was carried out on an estate with no deer present (Lochindorb) and the research was criticised for potentially confounding treatments, lack of replication and no meaningful experimental control. A recent study found no effects of Mountain Hare abundance on grouse tick burdens and actually found better grouse chick survival in areas with greater numbers of Mountain Hares, although tick biting rates were low on both species. There is a joint SNH-GWCT-SLE statement on voluntary restraint that acknowledges the lack of evidence to support population control of Mountain Hares to benefit Red Grouse. Despite all of this, the reporting of disease transmission as a primary ‘reason for shooting Mountain Hares’ was still more widely stated than any other reason in the questionnaire responses to the Review Group, and the advice that Mountain Hare culling should be part of tick disease control strategies is still present on some key land management websites (e.g. https://www.gwct.org.uk/policy/position-statements/mountain-hare-management/).

Reducing parasite burden on Mountain Hares and risk of wildlife diseases

The little information available reveals no evidence that parasite burdens on Mountain Hares are affected by their population density. Very little is also known about the risks of wildlife diseases (such as RDHV2) that could affect Mountain Hare populations in Scotland, but as yet do not appear to be doing so.

In addition to the above, it has been variously stated that reducing Mountain Hare numbers could increase predator pressure on other prey species (including Red Grouse), but there is no substantive evidence to support or refute this.

Medicated grit

Background

The presence of the strongyle worm (Trichostrongylus tenuis) in the gut of Red Grouse can cause cyclical fluctuations in grouse numbers every 6-9 years in Scotland. The use of 3-8 mm medicated grit in the form of quartz grit coated with a worming agent (currently flubendazole) has substantially suppressed these grouse cycles. The medicated grit is delivered to individual birds via grit stations deployed across the moor and the dosage should be based on pre-determined worm levels in the grouse. The delivery of flubendazole-coated grit to individual birds is overseen by local veterinary surgeons who can prescribe medicated grit and, when appropriate, advise on gritting holidays when the medicated grit is replaced in the tray by uncoated quartz grit.

Success following the introduction of flubendazole-coated grit can be judged by strongyle worm burdens now registering all-time lows, and by an almost doubling of grouse densities on driven grouse moors since its introduction in 2007. Given that previous efforts at controlling the strongyle worm without medicated grit were unsuccessful, and ever-increasing pressure on grouse by protected predators (as evidenced from the Langholm Moor projects), its use is now deemed essential if grouse levels are to be kept high enough to yield a harvestable surplus. Even with low worm burdens, managers may treat grouse as a precautionary measure, feeling it is the only mortality factor they can control, against the many other uncontrollable risks that grouse face. Consistent grouse numbers, year on year, help maintain viable levels of shooting, underpin investment decisions and contribute to the economic viability of grouse moors. Flubendazole is also commonly used to treat worm burdens in sheep and cattle.

The principal legislation influencing the use of medicated grit includes the Veterinary Medicines Regulations 2013 with Guidance Note 13 on the use of Cascade in terms of prescription (need and use) and the Wildlife & Countryside Act 1981 (as amended) with regards to operations requiring consent on designated sites. The amendments to the 1981 Act introduced by the Wildlife & Natural Environment Act (Scotland) Act 2011 also relate to direct dosing of grouse, which we do not consider here, rather than the use of medicated grit. Guidance on the use of medicated grit is provided in the Moorland Management Best Practice Worm Control in Red Grouse Guidance (Scotland’s Moorland Forum) and Best practice use of medicated grit (GWCT). Not all estates determine worm burdens and some use medicated grit as an insurance. Even for estates that do determine worm burdens, the evidence is not always given sufficient weight and moors continue to treat worms against the perceived wider risks to the grouse population from weather, predators and tick-borne disease.

Review of evidence

Having reviewed the written and verbal evidence presented to the Group we find that, when used correctly, flubendazole is a widespread treatment that has proved highly effective in reducing endemic strongyle worm levels in grouse guts. Also when used correctly, its residues in grouse for human consumption currently appear to present a very low risk. As past efforts of controlling strongyle worm burdens have not worked, the use of medicated grit is a key factor in maintaining consistent grouse numbers year on year.

The dosage of medicated grit supplied to birds must be determined by a veterinary prescription but this should reflect the current worm burden in the grouse in terms of health and breeding success. Whilst veterinary surgeons control the overall dosage, land managers determine the actual delivery of medicated grit to individual birds on the hill via grit trays, acknowledging that under the Cascade system, the prescribing veterinary surgeon must be satisfied that the individual using the medicated grit does so correctly. The prescribing veterinary surgeon has responsibility for the health of the grouse and should have seen the grouse before prescribing. Good practice includes testing a selection of birds to determine their worm burden. Low burdens indicate that gritting holidays could be taken, balancing this against the known risk of slightly lower grouse breeding success. Land managers are also required by law to ensure that no drug is ingested 28 days before the grouse are harvested. Medicated grit is commonly delivered in a grit tray with a lid to prevent access during that period. At present, there is little evidence for resistance to flubendazole developing in the strongyle worm population, but the main purpose of ‘gritting holidays’ is to reduce the chance of such resistance developing.

In terms of contested evidence, there are concerns that prescription levels when measured against the worm burden are too high, that gritting holidays reflecting low worm burdens are not always observed, and that grit may not always be withdrawn from grouse at least 28 days before Red Grouse enter the food chain. Although attempts have been made to address these issues in terms of Scotland’s Moorland Form’s Worm Control in Red Grouse Guidance and GWCT’s Best practice use of medicated grit, plus recent workshops provided by the Moredun Research Institute, there is no system in place for monitoring the use of medicated grit. Although there are estates following sustainable practices in the use of medicated grit, many are not. There is a need for estates to address this and bring the standards up to best practice usage as described in the Scotland’s Moorland Forum and GWCT best practice guidance.

We also identify a number of significant evidence gaps. There is anecdotal evidence of grit piles on open mounds instead of in bespoke grit trays, and similar claims of grit piles being found too near to water courses. Both these situations could lead to greater environmental contamination. Although there is as yet little evidence of a resistance problem with the use of medicated grit, more research is required on the potential development of such resistance and its implications. Better calibrated dosage reflecting worm risk based on weather and worm burden (determined by autumn worm counts and spring worm egg counts) would help allay the risk of resistance developing. There is some evidence that flubendazole is toxic to aquatic organisms; accordingly GWCT guidance recommends that grit trays be located no closer than 5m to running or standing waters. At present SEPA does not test for the presence of flubendazole contamination in water bodies. Current testing for flubendazole residues in the food chain is based on a small number of grouse samples each year and it is unclear what level of contamination would constitute a threat to human health. Although Cryptosporidium has only been detected once on grouse moors in Scotland, its presence in Northern England and connection to increased densities of grouse, with higher numbers visiting each grit tray, raises concerns for the future.