The management of wild deer in Scotland: Deer Working Group report

The final report of the Deer Working Group.

Section 15 Public Safety

1 The standards of public safety that should apply to deer management in all circumstances, including those related to shooter competence and food safety, were considered in Part Two of the Report. This Section considers the damage or risk of damage to public safety that can arise in particular circumstances.

2 The major issue is road traffic accidents involving deer. These ‘deer vehicle collisions’ (DVCs) are continuing to increase and are the subject of the rest of the Section.[1] The Group recognises that there can be other incidences involving public safety, such as the need to control deer on or near airport runways.[2] However, those cases are relatively isolated and not considered further here.

15.1 Frequency and Distribution of DVCs

3 Road traffic accidents (RTAs) involving wild deer have become increasingly recognised as a major public safety concern over the last 25 years. These accidents are generally referred to as DVCs and include both cases where there is a collision with a deer and cases where a vehicle swerves to avoid a deer and collides with something else.[3]

4 Following the inclusion of public safety in the Deer (Scotland) Act 1996 and the appointment of the first Deer Commission for Scotland (DCS) Board under the Act at the end of 1998, the DCS commissioned its first research considering DVCs in 1999.[4] The subsequent report in 2000 led to the DCS commissioning from 2003 the ‘DVCs in Scotland: Monitoring Project’ to collect and analyse information on the frequency and distribution of DVCs.[5] This project, which has been continued by Scottish Natural Heritage (SNH) since 2010, provides a systematic sample of DVCs in Scotland from 2003 to the present, with the data since 2008 more directly comparable year to year in terms of coverage and consistency of approach to data collection.

5 The monitoring project analyses data on DVCs from four main sources: the Trunk Road Operating Companies (TROCs) that manage the trunk road and motorway network for Transport Scotland; the police, for DVCs involving human injury; the Scottish Society for the Prevention of Cruelty to Animals (SSPCA), who are often called out when there are wounded deer; and Forestry and Land Scotland (FLS) rangers, who may be called out for injured deer near National Forest Estate land. Figure 33 shows the numbers of DVCs reported by these and other sources from 2008-18.[6] Figure 34 also illustrates the distribution of the recorded DVCs from the four main sources.

Figure 33 Number of DVCs recorded in the core data sources (2008-2018)
Year TROCs Police FLS SSPCA Others Total
2008 480 88 62 319 186 1,135
2009 652 75 101 291 425 1,544
2010 717 64 68 349 317 1,515
2011 593 72 104 419 23 1,211
2012 745 74 84 666 25 1,594
Mean 2008 - 2012 637 75 84 409 195 1,400
2013 638 81 73 698 94 1,584
2014 674 47 76 475 65 1,337
2015 660 30 62 883 63 1,698
2016 672 24 36 1,001 64 1,797
2017 620 30 48 1,255 53 2,006
Mean 2013 - 2017 653 42 59 862 68 1,684
2018 530 TBC 56 1,102 68 1,756
Total 6,981 585 770 7,458 1,383 17,177

Source: Langbein (2019)

Figure 34 Distribution and frequency of DVCs per 4km by 4km square from all core data sources
Figure 34 Distribution and frequency of DVCs per 4km by 4km square from all core data sources

6 These records are only a sample of the DVCs in Scotland. However, consistency in data collection means the records can provide indicators of changes in frequency and distribution. This helps identify areas with high frequencies of DVCs or where incidences are increasing. The monitoring focuses on the main strategic roads in Scotland, where traffic is concentrated. Trunk roads (motorways and major strategic A-roads), for example, account for 6% of the road network and yet carry 39% of all traffic and 63% of HGVs.[7] These trunk roads, combined with non-trunk A-roads, account for around 75% of the DVC records, while representing around 20% of the road network.[8]

7 The patterns of DVCs recorded across the road network were similar in the periods 2008-12 and 2013-17, but with a 20% increase in the number of reported DVCs in the second period. There are not many species-specific records. Those that do exist show that red deer are regularly involved in a significant proportion of reported DVCs in five Local Authority areas, particularly Highland and Argyll and Bute. Sika are reported in a small proportion of DVCs in those Local Authority areas, while records of fallow mostly relate to Dumfries and Galloway and Perth and Kinross. For the majority of other Local Authority areas, roe deer are the only species recorded as involved in DVCs.

8 The approach that has been used by the project to monitor DVCs was not intended to provide estimates for the total number of DVCs in Scotland each year, and there are many variables to consider when trying to extrapolate from the current data to make national estimates. These include, for example, the lack of clear information on the proportion of all DVCs on the lengths covered by TROCs that are recorded by the companies.[9] As a result of such factors, a national estimate for DVCs in Scotland in 2017 was given as being between the broad margins of 4,000-12,000 per year.[10]

9 Estimates at a UK level have suggested that there are more than 450 DVCs a year involving human injury, with 10-20 fatalities.[11] While 70 of those DVCs were estimated to be in Scotland, it is considered that the actual number of DVCs in Scotland involving human injury may exceed 120 per year.[12] This equates to 1% or more of the estimated number of DVCs. A review of the available police Personal Injury Accident (PIA) records in Scotland in which deer have been implicated, suggests that nearly 60% of the cases involved a driver swerving and hitting another car or other object, rather than a direct collision with a deer.[13]

10 In the past, there appear to have been difficulties obtaining PIA records from all the police forces across Scotland and it is anticipated that this situation might improve now there is a single force, Police Scotland. However, a limitation to the value of the PIA records in this context, is that the Department of Transport form used by Police Scotland (the ST19 form) does not have a separate category for deer. Attempts to change this since the time of the DCS have been unsuccessful.[14] The Group considers that Transport Scotland and SNH should be continuing to push for this change on ST19 forms used in Scotland, and that the Scottish Government should directly support that.

11 The Working Group recommends that the Scottish Government should be working to ensure that the UK Department of Transport form used by Police Scotland to record Personal Injury Accidents (ST19), is modified for use in Scotland to include a separate category for deer.

12 A DVC where nobody is injured can also be a fairly traumatic experience for those involved. There have been indications previously that the risk of being involved in a DVC could be twice as high per driven mile in Scotland than in England and Wales.[15] The position may become clearer as SNH has recently commissioned research using the existing DVC records to assess the risk of a DVC per road length.[16]

13 SNH is also continuing the DVC monitoring project, with data from the 2016-18 phase indicating that the number of DVCs in Scotland is continuing to increase.[17] This has led to a revised national estimate of “8,000 up to 14,000” per year in Scotland.[18] The Group supports the monitoring project continuing to develop through further phases. The project is jointly funded by SNH and Transport Scotland. The Group did not examine whether the balance between SNH and Transport Scotland in funding this project and some other research related to DVCs is appropriate, given Transport Scotland’s more direct responsibilities for road safety.

14 The Group considers that there is a clear logic in the DVC monitoring project being focused on the motorway and trunk road network, as it carries a disproportionate amount of Scotland’s traffic. However, that means 75% of the DVC records come from 20% of the road network and there is a lack of systematic data on the frequency and distribution of DVCs over the rest of the road network under Local Authority management.

15 The Group recognises that SNH is engaging Local Authorities on this topic, as part of assisting Local Authorities to produce Deer Management Plans for their areas. However, the Group considers that more progress is needed on identifying DVC ‘hotspots’ on roads managed by Local Authorities and then considering the most appropriate mitigating measures for those road lengths.

16 The broad range quoted in paragraph 13 for the possible total number of DVCs in Scotland each year reflects, as discussed above, that the DVC monitoring project was never intended to be able to give national estimates and is unsuited to doing so. The Group recognises that trying to obtain an overall total of the number of actual DVCs is unrealistic. However, the Group considers that having a clearer estimate would be helpful in ensuring there is an appropriate allocation of resources to reduce the frequency of DVCs.

17 A proposal has been made that a clearer national estimate could potentially be obtained through a one-off independent public questionnaire, with the results combined with the existing DVC data to produce a better estimate of the true total.[19] The Group is not qualified to judge the merits of this proposal and how it might be carried out, but the Group considers that it should be actively considered.

15.2 Costs of DVCs

18 Thousands of people in Scotland are involved in DVCs in each year. This represents the most direct and common adverse impact of wild deer on people. DVCs also appear to be the largest economic cost of wild deer compared to other forms of damage. A study in 2012 estimated that DVCs in Scotland cost £9.4 million and, in 2016, SNH revised that estimate to £13.8 million.[20] The Group considers that improved information would be likely to show that the actual costs are even higher.

19 There are a range of direct and indirect costs involved in DVCs. The direct costs include the human injuries and any fatalities. The costs of these are conventionally calculated using the Department of Transport’s figures for the value of preventing accidents.[21] In 2012, these figures gave average values for all types of roads of £1.92 million per fatality, £219,043 per serious injury and £23,336 per minor injury.

20 The figures for the number of accidents involving fatalities or injuries each year that are central to calculating these values, are based on the information in the ST19 forms completed by the police. The Department recognises the number of cases involving injuries is an underestimate as the police are not always involved in such incidences.[22] However, the particular significance of the values is that they can form the basis for Transport Scotland justifying expenditure on mitigating measures along stretches of road to reduce DVCs.

21 Another direct cost of DVCs is damage to vehicles. The AA estimated in 2009 that insurance claims in the UK for DVCs could amount to between £59 million and £103 million a year.[23] While no separate figures are available for Scotland, around 20% of DVCs are estimated to be in Scotland and the insurance costs will run to millions of pounds. In addition, not all the DVCs involving vehicle damage will result in an insurance claim.[24] Other direct costs of DVCs include the time of those involved in dealing with DVCs including: injured deer; the costs of collecting and disposing of carcases; DVC research and monitoring; and carrying out mitigation measures.

22 DVCs can also include indirect costs for people as a consequence of being directly involved in one, as well as for other people if a DVC results in a temporary road closure. There are well established methods for quantifying the value of the factors such as time lost and increased vehicle operating costs because of road delays. These were used in a case study in 2009 based on a scenario where a DVC caused a four hour road closure on a main Scottish route. The study concluded that the DVC would have resulted in quantifiable indirect costs of at least £75,000.[25]

23 The annual cost of DVCs in Scotland of £13.8 million estimated by SNH is a substantial sum and the Group considers that the actual costs are most likely to be even higher. The Group also considers that, while there has been increasing attention paid to DVCs over recent years, it is an issue which has yet to receive the level of attention that it warrants.

24 The Working Group recommends that the Scottish Government should ensure that a more appropriate level of attention and resources is applied to addressing the continuing rise in road traffic accidents in Scotland involving wild deer.

15.3 DVC Mitigation Measures

25 The main statutory responsibilities for trying to reduce DVCs to improve public safety are held by SNH, with its responsibilities for the management of deer, and by Transport Scotland and Local Authorities, with their responsibilities for the management of roads. SNH’s involvement includes its scope to carry out research and provide advice under s.3 of the Deer (Scotland) Act 1996, as well as to use its regulatory powers in the Act to control deer numbers.

26 SNH, like the DCS before it, has played an active part in raising awareness of DVCs as an issue. This includes advice to TROCs on the annual Deer Management Plans (DMPs) that they are now required to produce under their operating contracts.[26] This process is still in its early stages, but it is hoped that these will result in improved data collection by the companies and lead to a greater focus on identifying ‘hotspots’ and applying appropriate mitigating measures.

27 SNH also engages Local Authorities over DVCs, both in advising on the DMPs that the Authorities are now committed to producing and through local Community Planning Partnerships. SNH also raises the issue with land managers, for example through the DMPs that are part of the current DMG assessment process.[27] More generally, the Code of Practice on Deer Management encourages land managers to recognise road safety as a public interest when planning deer management and to “contribute to co-ordinated action to reduce road safety risks”.[28]

28 As part of raising the awareness of drivers to the risk of DVCs, SNH and Transport Scotland also run two campaigns a year using the Variable Message Signs on the trunk road network.[29] A campaign in the autumn starting on 30th October targets upland red deer areas to warn of an increased risk of red deer on roads as a result of seasonal movements associated with the rut. A spring campaign in late May/early June targets lowland roe deer areas to warn of an increased risk of deer on the road as a result of the dispersal of juvenile roe deer.

29 There has long been a general and understandable acceptance that the direct application of mitigation measures should be focused on the ‘hotspots’ where DVCs are most frequent.[30] Prior to SNH becoming responsible for the deer legislation, the DCS had used its s.10 powers to assist reductions in deer numbers in a couple of priority locations for road safety.

30 The DCS also initiated a project in 2009/10 with Transport Scotland to investigate the effectiveness of mitigating measures at priority sites along three roads (A835 Garve, A87 Lochshiel, A82 Glencoe).[31] For each site, the DCS appointed a Panel under s.4 of the 1996 Act to advise on the measures to be undertaken. These included the use of vehicle activated warning signs, some vegetation clearance along the roadside and some culling. The results included a significant reduction in the number of carcases on the A82 and A87, but not the A835. There was, however, a general reduction in drivers’ speeds and the number of DVCs.

31 The DCS project illustrated some of the range of factors that can influence the frequency of DVCs on particular stretches of road, including local deer densities, verge management, the adjoining vegetation, lines of sight, driver awareness and vehicle speed. The main measures to apply in any situation will depend on the particular context.

32 On motorways and high-speed trunk roads, deer fencing along the highway remains the most effective measure against DVCs. In upland red deer areas in particular there may also still be a need to allow lengths where the deer can cross as part of their seasonal movement. The involvement of SNH in providing advice as part of large-scale road schemes, such as the M80 upgrade and dualling of the A9, allows such factors to be considered at the planning stage, including the use of underpasses for deer movement and other wildlife.

33 There is extensive experience in Europe in using under- and over-passes and other measures to reduce the risks of collisions with deer and wild boar, as well as bears in some areas and elk in Scandinavia. The need for research in this country “to test different mitigation options in order to keep abreast of technological advances and successful mitigation systems being used in Europe” is recognised by SNH and Transport Scotland.[32]

34 Research has previously indicated that some types of measures are not effective at reducing DVCs, such as roadside reflectors to reflect headlights along the verge at night and high frequency whistles attached to vehicles. Drivers also get used to static deer warning signs and ignore them, while clearing woodland or scrub back from the verge to improve lines of sight needs to avoid creating a grass strip that attracts deer to feed.

35 A current research proposal being considered by Transport Scotland and SNH involves testing two types of mitigation measures for the trunk road network in Scotland.[33] One test involves roadside devices which emit strobe light and variable acoustic signals into the verge when triggered by approaching vehicles. The other test involves vehicle activated signs that display vehicle speed and a deer warning, with existing evidence indicating that reducing vehicle speed can be an important factor. Both tests were to be carried out at a number of sites with high frequencies of DVCs.

36 The Group considers that research on different types of mitigation measures is an important contribution to trying to reduce DVCs. The Group also considers that, as part of that, more attention should be paid to local deer densities in the areas around ‘hotspots’. It has long been recognised that there is a relationship between the frequency of DVCs and deer densities, and that reducing densities could reduce DVCs.[34]

37 There appear to have been surprisingly few studies of the influence of reducing deer densities. Those that exist indicate that, as would be expected, there is not a linear relationship between reduced deer densities and reducing DVCs due to the types of other variables mentioned above relating to the nature of the particular stretch of road involved. However, DVCs are positively correlated with deer densities and the Group considers that local deer densities are a very important factor to be taken into account.[35]

38 There is no evidence that the presence of more deer in an area is likely to reduce DVCs and, even if reducing densities in some areas might not result in a demonstrably direct reduction in DVCs, controlling adjacent deer densities should be a basic component of mitigating measures to reduce DVCs in recognised ‘hotspots’.

39 In most parts of Scotland, roe deer are either the only deer species causing DVCs or the main species. The Group considers that particular attention should be focused on the densities of roe deer in wooded areas along roads with frequent DVCs in more lowland environments, including peri-urban areas.

40 The clear indications are that woodlands often have high densities of roe deer due to under-culling and that one of the consequences of those densities is higher levels of dispersal, leading to increased risk of DVCs in places. The Group considers that land managers are often not aware of the cull levels required to contain the size of a roe deer population in lowland environments, where a roe doe population may have an average 100% calving rate.

41 The Group considers that SNH should be more active in using s.40 cull returns to establish the current culling patterns in a corridor down either side of lengths of road that have been identified as ‘hotspots’ for DVCs involving roe deer. SNH can also assess the relative densities of the roe deer from their impacts on the vegetation in the woodland, while monitoring DVCs with the TROC or Local Authority responsible for the road. SNH can then promote cull levels that ensure the densities of deer are both within the capacity of the habitats and produce a reduction in DVCs.

42 The Group considers that the fact that there tends not to be linear relationship between deer densities and DVCs due to the other factors, can result in not enough attention being paid to local deer densities. The Group’s understanding is that there are no roads in Scotland where SNH is taking a systematic approach to the control of deer densities along lengths that have frequent DVCs.[36]

43 The Group considers that, while a more focused approach to local deer densities should be used along some lengths of trunk roads,[37] the management of local deer densities is likely to be a key approach along more minor roads. As things stand, there is unlikely to be much coverage of these roads by Variable Message Signs or other relatively expensive devices, while installing and maintaining deer fences is also expensive. Managing local deer densities therefore becomes relatively more important as a potential mitigating measure that can be carried out locally to try to limit the frequency of DVCs.

44 The Working Group recommends that Scottish Natural Heritage should be paying much more attention to the control of local deer densities alongside lengths of public roads with frequent road traffic accidents involving wild deer.

45 More generally, the Group considers that improved information about DVCs and the need to try to reduce DVCs for public safety and wider economic reasons, should become an increasingly important influence on local deer management in many areas.


1 There are also deer-train collisions but they are not considered a risk to public safety. Network Rail has records of around 500 deer-train collisions in Scotland between 2008-14, but these records are unmapped and there is no data on the costs involved in deer-train collisions (Langbein Wildlife Associates, Phase I Preliminary desktop review to assess the scale and distribution of past deer-vehicle collisions and identify priority areas for field survey, A9 Dualling Programme, Central Section Glen Garry to Dalraddy, December 2015).

2 For example, the use of s.10 powers at Machrihanish, as mentioned later in Section 23 Emergency Control Measures.

3 SNH (2016). Deer Management in Scotland: Report to the Scottish Government from SNH.

4 Staines, B. (2000). Wild Deer issues concerned with deer welfare and public safety. Deer Commission for Scotland.

5 Langbein,J. (2017). Deer-vehicle collisions in Scotland: data collection and collation to end of 2015. SNH Commissioned Report
No. 950.

6 Langbein, J. (2019). Deer-Vehicle Collision (DVC) Data Collection and Analysis to end 2018. Scottish Natural Heritage Research Report.

7 Transport Scotland, Scottish Transport Statistics.

8 Langbein, J. and Putman, R. (2006). National Deer-Vehicle Collisions Project: Scotland 2003-2005. Final report to Scottish Executive. The Deer Initiative, Wrexham.

9 Langbein (2017) Op cit.

10 Langbein (2017) Op cit.

11 Langbein (2017) Op cit.

12 Langbein (2017) Op cit.

13 Langbein (2017) Op cit.

14 For example, see DCS Annual Report, 2000/01.

15 Langbein, J. (2008). DVCs in peri-urban areas: a risky life for deer. Presentation at the British Deer Society conference, London.

16 SNH Information Response 10.

17 Langbein (2017) Op cit.

18 Langbein, J. (2018). Deer-vehicle collisions in Scotland: National data overview, regional hotspots, trends, mitigation. Presentation to Lowland Deer Network, June 2018.

19 Langbein (2017) Op cit.

20 Putman, R. (2012). Scoping the economic benefits and costs of wild deer and their management in Scotland, SNH Commissioned Report No. 526; and SNH (2016) Op cit.

21 Department of Transport (2012). A valuation of road accidents and casualties in Great Britain in 2012.

22 Department of Transport (2012) Op cit.

23 The AA (2009). ‘Deer Collision Claims: Autumn rut costs drivers deer’, AA website, 30 September.

24 Langbein (2017) Op cit.

25 JMP Consultants Ltd (2009). The Indirect Costs of Deer Vehicle Collisions. Report to SNH, 31 March.

26 In accordance with Schedule 7 (Part 4) of the annual landscape management report required as part of the fourth Generation Team Contract for the management and maintenance of the Scottish Trunk Road Network.

27 See Section 26.

28 SNH (2001). Code of Practice on Deer Management, p.14.

29 SNH Information Response 10.

30 White, P.C.L., Smart, J.C.R., Böhm, M., Langbein, J. and Ward, A.I. (2004). Economic impacts of wild deer in the East of England. Report for the Forestry Commission and English Nature.

31 Transport Scotland/Scotland Transerv (2010). North West Area A82, A87, A835 Vehicle Activated Deer Warning Signs, Report 07/NW/0805/046.

32 SNH Information Response 10.

33 SNH Information Response 10.

34 For example, DCS Annual Report, 2003/04.

35 SNH (2016) Op cit.

36 DWG meeting with SNH, 19 June 2019.

37 For example, along parts of the A9 in northern Perth and Kinross where high frequencies of DVCs occur.



Back to top