Scottish Natural Heritage Commissioned Report No. 948 Trends in woodland deer abundance across Scotland: 2001-2016 C O M M I S S I O N E D R E P O R T Commissioned Report No. 948 Trends in woodland deer abundance across Scotland: 2001-2016 For further information on this report please contact: Dominic Sargent Scottish Natural Heritage Torlundy FORT WILLIAM PH33 6SW Telephone: 01397 704716 E-mail: [email protected] This report should be quoted as: Campbell, D., Marchbank, M., Watson, M. & Quin, S. 2017. Trends in woodland deer abundance across Scotland: 2001-2016. Scottish Natural Heritage Commissioned Report No. 948. This report, or any part of it, should not be reproduced without the permission of Scottish Natural Heritage. This permission will not be withheld unreasonably. The views expressed by the author(s) of this report should not be taken as the views and policies of Scottish Natural Heritage. © Scottish Natural Heritage Year 2017. COMMISSIONED REPORT Summary Trends in woodland deer abundance across Scotland: 2001-2016 Commissioned Report No. 948 Project No: 016720 Contractor: Strath Caulaidh Ltd. Year of publication: 2017 Keywords Wild deer; Scotland; woodland; population modelling; historic trends; abundance; density. Background Prior to the recent review of deer management by SNH, it became evident that contemporary estimates of wild deer abundance at a national scale were not available. SNH contacted Forest Enterprise Scotland (FES), which manages the National Forest Estate (NFE), to enquire if their national monitoring data could be compiled and analysed to provide a summary of spatial and temporal trends in woodland deer densities across the NFE. The findings could then be included in the SNH review report for Ministers. Deer population models were built for 32 forests representing a cross section of conditions across Scotland within the NFE, and covering 38% of the land area under management. Trends in deer abundance over the period June 2001 to June 2016 in each forest were predicted from the models, which used historic deer cull records, estimates of historic deer recruitment rates and estimates of abundance obtained from routine monitoring of the NFE using deer dung counts. The estimates of deer abundance for June 2016 from the 32 forests modelled were used in an extrapolation exercise which produced an estimate of the total number of deer considered to be currently present within the NFE as a whole. Using national cull records provided by SNH an estimate of the number of deer present in private woodlands was also produced, albeit using a much less complex methodology, so that a broad contemporary estimate of overall woodland deer abundance for Scotland could also be obtained. The main findings of this review were incorporated in the SNH review of deer management presented to Ministers. Main findings  The results of the population modelling exercise indicate that deer population density at the national scale on the NFE has declined markedly since June 2001, albeit this general trend masks a wide range of site-specific trends (decreasing, stable and increasing population density at the local scale). i  An overall decline in deer numbers of between 20% and 30% has evidently occurred between June 2001 and June 2016, but with the majority of the change occurring in the past five years.  The modelling indicates the abundance of wild deer on the NFE in June 2001 may have been ~ 109,000 or 16.5 deer per km2. On the basis of culls taken, and prevailing recruitment rates over a 15 year period, that number is believed to have fallen by June 2016 to ~ 82,000 or 12.5 per km2. This represents a decline of 24% or ~ 26,500 deer over a 15 year period.  When the overall trend is broken down into regions, it appears that deer densities over the 15 year period have declined markedly more in the South and in the North than in the Central operations areas of the NFE. Moreover, roe/sika/fallow deer densities appear to have declined markedly more than red deer densities nationally across the NFE.  A complex variety of interacting factors appears to be responsible for the temporal and spatial trends observed in each region, and between deer species groups. These include: recent differences in culling intensity between regions; inherent differences in the proportions of each deer species present (and resulting differences in overall recruitment, as each species recruits at different rates); differences in the type of deer controllers used (employed versus contract); differences in weather patterns; and variation in the degree of external influences apparent (e.g. red deer densities in woodland are in part controlled by factors such as migration from open range populations, which is the likely reason why they have not declined as quickly as roe/sika/fallow over the same period).  The modelling exercise for the NFE as a whole indicates that the current likely size of the population in June 2016 is 40,000-45,000 roe/sika/fallow deer and 40,000-45,000 red deer (total of 80,000-90,000 deer).  The most recent annual culls on the NFE have averaged ~ 30,000 – 32,000 animals. If this level of culling continues across the NFE as a whole it is likely that population densities, on average, will continue to decline. That said, there is also good evidence to show that densities are currently rising locally in a proportion of monitored forests, notably so in terms of red deer populations. Allocation of additional culling effort to these places would be required if a consistently low density population of deer is to be achieved, as is the aim of FES nationally so that all their key crop and habitat management targets can be met.  There is good evidence to show that recruitment rates are rising in many places on the NFE as a result of declining deer densities and, perhaps, due to continuing changes in forest structure. Therefore, culling levels will need to take this into account in future alongside consideration of other key practical factors including the implications of tree species changes due to disease (implications for carrying capacity), declining budgets for forest operations generally and the continuing movement of deer from private land onto the NFE locally (e.g. over the 15 year period since June 2001 it is estimated that ~ 16,000 male red deer and ~ 5,000 male sika deer shot on the NFE migrated into NFE forests from neighbouring private land).  The situation in private woodlands is much less well understood, even with the various elements of analysis undertaken for this report. The various analyses show that deer culling intensities are lower than on the NFE and recruitment rates are also lower (normally a sign that deer are at higher density, given the geographic distribution of sites and the habitat in private woodlands overall are both broadly similar to that of the NFE). Much more data gathering and analysis is required if the effectiveness of deer management in these areas is to be quantified. In the interim, the results of this study indicate it is safest to assume that deer densities on average are currently somewhat higher in private woodlands than on the NFE, and that populations in general are more likely to be stable on average than declining (as opposed to declining markedly overall as on the NFE). ii  The last section of this report makes a range of recommendations on possible ways to improve our understanding of spatial and temporal trends in wild deer abundance in woodland habitats at the national scale in the future. For further information on this project contact: Dominic Sargent, Scottish Natural Heritage, Torlundy, Fort William, PH33 6SW. Tel: 01397 704716 or [email protected] For further information on the SNH Research & Technical Support Programme contact: Knowledge & Information Unit, Scottish Natural Heritage, Great Glen House, Inverness, IV3 8NW. Tel: 01463 725000 or [email protected] iii Table of Contents Page 1.  INTRODUCTION 1  1.1  Study background 1  1.2  Aims of this study 1  2.  THE NATIONAL FOREST ESTATE 2  2.1  Overview 2  2.2  Monitoring approach 5  2.2.1  Impacts 5  2.2.2  Culls taken 6  2.2.3  Deer density 7  3.  METHODS 9  3.1  Study Site Selection 9  3.2  Models of deer population dynamics 10  3.3  Compilation & extrapolation of model outputs 13  3.4  Factors influencing abundance trends on the NFE 15  3.5  Abundance trends in private woodlands 16  3.6  Comparison with open range populations 16  4.  FINDINGS 17  4.1  Modelled abundance: part of the NFE 17  4.2  Predicted abundance: the entire NFE 24  4.3  Factors influencing trends on the NFE 32  4.3.1  Patterns of culling 32  4.3.2  Culling method 33  4.3.3  Patterns of recruitment 34  4.3.4  Underlying changes in forest structure 36  4.3.5  Winter weather and site access 37  4.3.6  Other factors 37  4.4  Predicted abundance: private woodland 38  4.5  Predicted abundance: all woodlands 43  5.  CONCLUSIONS 45  6.  RECOMMENDATIONS 46  ANNEX 1: POPULATION MODEL OUTPUT AND CULL RECORDS 47  iv Report preparation Contributors  Douglas Campbell  Mel Marchbank  Matt Watson  Sam Quin Draft by / date Checked / date Final by / date Checked / date MM 15/06/16 DC 20/06/16 SQ 05/08/16 DC 10/08/16 Standard caveats  SCL have exercised reasonable skill, care and diligence in the preparation of this document, in accordance with the standards of a qualified and competent person experienced in carrying out work of a similar scope and complexity to the agreed services and current at the time when the services were performed.  SCL have performed the agreed services generally in accordance with our proposal document or otherwise according to the clients specification, but have in places added to and varied the scope where it appeared to us necessary and reasonable to do so.  SCL have taken all reasonable precautions to avoid damage to property belonging to the client and any third party.  SCL worked with sub-contractors to perform part of the services and we exercised all reasonable care to ensure that they were appropriately skilled and experienced in relation to the work that they were instructed to carry out.  The services and the service products delivered to date cannot necessarily reveal all adverse or other material conditions at the site that could otherwise be identified either through a different formulation of the services or through more detailed work being carried out by SCL. Specific caveats  The report in places uses data sets created by other organisations and we cannot be held responsible for their accuracy.  The timeline for this project was very short, and as a result the full range of analysis we would have liked to undertake was not possible. The Interpretation section of this report lists the related issues and describes potential resolutions. v Acknowledgements This report was prepared by Strath Caulaidh Ltd with contributions from the following external organisations:  Ian Fergusson of Forest Enterprise Scotland (FES) for supplying cull records  Donald Fraser of Scottish Natural Heritage (SNH) for supplying cull records  Bruce Sewell of FES and Chris Nevin of SNH for jointly locating the funding for the study vi 1. INTRODUCTION 1.1 Study background The Scottish Government (SG) recently undertook a review of deer management. When evidence was taken by the Rural Affairs, Climate Change and Environment Committee overseeing the review, it became evident that contemporary estimates of wild deer abundance at a national scale were not available. Scottish Natural Heritage (SNH) is currently preparing a report on progress in deer management for Scottish Ministers. Part of that report includes information on deer populations and their trends. One relevant source of information is woodland deer density records for the National Forest Estate (NFE). SNH contacted Forest Enterprise Scotland (FES), which manages the NFE, to ask whether the data could be compiled and analysed to provide a summary of spatial and temporal trends in deer densities nationally in NFE woodlands. The findings could then be fed into the report for the Scottish Ministers as part of the ongoing review of deer management in Scotland. Strath Caulaidh Ltd (SCL) has gathered all of the woodland deer density data available from NFE land, on contract, since the late 1990’s. FES and SNH jointly asked whether SCL, given its long historic involvement in the subject, would be able to help scope out and undertake the proposed study. 1.2 Aims of this study The final scope of work SCL agreed with SNH and FES was as follows:  Compile deer density survey data held for the NFE and summarise it. The time available for the project1 prevented all records from being compiled and analysed for this report, hence it was agreed SCL would provide data for a representative cross- section of the NFE (SCL proposed ideally ~ 50% of the land mass managed).  Obtain the full national cull records for the NFE for the past 15 years from FES, and analyse to produce statistics (recruitment rates and size of culls taken) for use in deer population models as well as for presentation purposes.  Input the selected density data sets, and statistics derived from the cull records, into long-term deer population models spanning an agreed period (15 years, from 2001- 2016). Use the models to forecast how many deer are likely to be present in each surveyed area in summer 2016. Model output will also be summarised to provide an overview of population trends over the past 15 years, in light of the size of culls taken relative to the rate of recruitment apparent, at the regional and national scales.  Prepare a short written report, containing chart and map-based output as required, that will describe: o The methods used. o The key findings of the study (surveyed deer densities, culls taken, prevailing recruitment rates and resulting population trends over a 15 year period). o An explanation of the observed trends. o Any key recommendations arising. 1 Two weeks were available to undertake this work, because of the overall deadline in place for the report to be produced. 1 2. THE NATIONAL FOREST ESTATE 2.1 Overview Forest Enterprise Scotland (FES) manages the National Forest Estate (NFE) in Scotland. The NFE comprises ~ 658,000ha of land spread across the Scottish mainland and islands (Map 1). The exact size of the landholding varies from year to year due to disposals and acquisitions. The individual properties range in size from less than 100ha to over 10,000ha. The most extensive contiguous landholdings are in Galloway, Argyll-shire and Sutherland. A large proportion of the NFE comprises commercial forest from which timber products are derived for national and international consumption. Some forests were established as early as the 1920’s for this purpose, but most were established from the 1960’s onwards; a small number of new forests continue to be created each year. The harvesting and re-stocking of these forests for timber production is a major part of the work undertaken by FES managers. However, many landholdings within the NFE also contain semi-natural woodland remnants along with planted broadleaf woodland. The most recent acquisitions tend to be for new native woodland establishment projects. Native woodland restoration is another major focus for FES managers. The NFE also contains a significant proportion of open range habitat, comprising a wide range of site types. Small areas of ride, streamside and glades are commonly found within all forests; a smaller number of landholdings comprise major mountain ranges which have more substantial tracts of heathland, grassland and bog present. Parts of the NFE comprise sites designated for nature conservation (a third major focus for FES managers), including a wide range of Sites of Special Scientific Interest (SSSI’s), Special Protection Areas (SPA’s) and Special Areas of Conservation (SAC’s). Four main species of wild deer are present on the NFE: roe (Capreolus capreolus), sika (Cervus nippon), fallow (Dama dama) and red deer (Cervus elaphus). Their grazing, browsing and trampling impacts on trees and ground vegetation can lead to problems in achieving management objectives across all site types if deer densities are inappropriately high. FES managers currently use a combination of two approaches to reduce deer impacts to an acceptable level: (i) control of deer by culling and (ii) protection using fencing. Culling tends to be undertaken across the entire landholding, the aim being to provide a broadly favourable environment for FES staff to re-stock trees, harvest timber and otherwise manage the environment to achieve conservation objectives. Fencing tends to be used for forest perimeters, most commonly in the red deer range, to prevent wild deer from neighbouring land entering the NFE. In some situations fencing is also used to protect individual planting sites. FES employs 45 in-house staff to control wild deer (‘Wildlife Rangers’ or WR) along with around 100 culling contractors working for FES and around 300 recreational stalkers who have Permissions to cull deer on the Estate. The individuals involved in culling are managed by Wildlife Ranger Managers (WRM’s) and in turn Deer Management Officers (DMO’s). The DMO’s are overseen nationally by the Forest Management Officer (FMO). The deer management operation on the NFE has gross costs of ~ £7 million per annum and net costs of ~ £5 million after £2 million revenues from leases, permit stalking and venison are accounted for. The argument in favour of a net cost in the deer department being sustainable is that otherwise the environment would rapidly become unfavourable for crop establishment (for example in 2015, 18.9% of Year 1 leading shoots on commercial crops 2