General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (2/05) Page 1 of 31

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

A SID 5A form must be completed where a project is paid on a monthly basis or against quarterly invoices. No SID 5A is required where payments are made at milestone points. When a SID 5A is required, no SID 5 form will be accepted without the accompanying SID 5A.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code SE3032

2. Project title

The demographic, ecological and epidemiological consequences of culling badgers

3. Contractororganisation(s)

Central Science LaboratorySand HuttonYorkYO41 1LZ                         

54. Total Defra project costs £ See Woodchester Park SID5a (SE3032)

5. Project: start date................ 01/04/2004

end date................. 31/03/2007

SID 5 (2/05) Page 2 of 31

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domainWe have no objection to this report being published but would appreciate the opportunity to have the time to check it through very carefully first to ensure that it does not contain any sensitive or confidential information, particularly with respect to data from the RBCT and the report from Sheffield University (Appendix 1).

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.

Social perturbation of badger populations following culling has been proposed as an explanation for the failure of culling to consistently demonstrate significant reductions in cattle TB. Field studies support this hypothesis by demonstrating an increase in badger ranging behaviour and an associated increase in the incidence of TB infection in cattle following culling. A greater understanding of the epidemiological consequences of social perturbation will be extremely useful to fully interpret the results of the RBCT and is required to enable the formulation of an effective strategy to control TB in cattle. This study compared the characteristics of a sample of badger groups in an area subjected to sequential culling over several years with that of an adjacent undisturbed population.

Main objectives

1. To obtain data on the spatial configuration of badger social groups subjected to culling.2. To collect data on the size, structure and infection status of a population subjected to culling.3. To collect data on the behaviour of individual badgers in response to culling operations.4. To obtain data on the efficiency of badger culling operations.

Methods and findings

An intensive programme of fieldwork began in the spring of 2004 within and adjacent to the proactive triplet I2 of the RBCT, two years after the initial culling operation and one year after the first follow-up cull. The study area covered 27 km2, approximately two thirds of which was inside the proactive area. Badger population size and structure was determined annually from a capture-mark-recapture programme between June 2004 and January 2007. During this period there were 244 captures in over 2,600 trap nights, accounting for 134 individual badgers. Clinical samples were collected from badgers at each capture event to investigate the epidemiological characteristics of the population. Population density was low throughout the area but was approximately twice as high in the unculled area than in the culled area. The spatial configuration of social groups was investigated by bait marking the entire area each spring between 2005 and 2007. This identified 22 discrete social groups, half of which were in the culled area. Individual ranging behaviour in response to culling was investigated by radio-tracking a sample of adult badgers from social groups in the culled area and the adjacent undisturbed area between June 2004 and December 2006. Approximately 1,350 daytime location fixes were collected on 40 badgers along with 3,000 radio fixes during more than 1,200 hours of nocturnal tracking.

SID 5 (2/05) Page 3 of 31

The mean social group range was almost three times larger in the culled area in 2005 than in the unculled area. In addition, the mean social group range in the culled area appeared to decrease dramatically after 2005. The distance between setts and their associated bait returns were on average 140 m further from setts in the culled area than the unculled area in 2005 and 2007.

Results from bait marking and radio-tracking indicated that badgers in the area subjected to culling ranged further than those in the adjacent undisturbed population. Bait marking indicated a distinct difference in the spatial organisation of the two populations that varied over time. Radio-tracking revealed that badgers in the area subjected to culling ranged more widely than those in the adjacent unculled area. This difference was most evident in their core ranges and increased with successive culling.

The ratio of cubs to adults was significantly higher in the culled area than the unculled area in 2005 and 2006 indicating increased productivity, possibly a compensatory density-dependent response to population reduction. The ratio of adult females to males was five times higher in the culled area than in the unculled area in 2005 indicating that the population had become more female biased. This may be partly due to female biased immigration into the culled area, which has been reported previously following culling. The reasons for the observed increase in the number of female cubs (relative to males) in the culled population are however unclear.

Evidence of trap-shyness was collected by monitoring sett activity levels prior to trapping. This suggested that badgers in the culled area were more neophobic because they were more likely to avoid areas where traps were set and were also less likely to be caught from active setts than badgers in the unculled area.

Results of the live-sampling indicated temporal and spatial differences in the prevalence of animals excreting M. bovis and also ELISA positive cases. In particular ELISA positives in the culled area dropped considerably from 39% to 5% in 2005 and then increased to 26% the year after. This pattern was repeated in the prevalence of culture positive cases determined from post mortem examination and tissue culture of badgers culled in 2004 and 2005. Critically, the peak in positive cases recorded in the unculled area in 2005 was concomitant with the most evident differences in ranging behaviour indicated by bait marking and radio-tracking.

Main conclusions

This study provides further evidence for culling-induced perturbation of badger social organisation. Uniquely however, these results relate to a population that was subjected to sequential annual culling and suggest that the process of demographic and individual disturbance persisted at least as long as culling continued. Although culling appeared to have a beneficial effect on levels of detectable infection in the culled area, infection levels in the adjacent area peaked when levels of social disturbance were greatest. These findings have direct relevance for assessing the sustainability of a range of policies to control bovine TB in badgers, and in agreement with previous studies suggest that culling may have both positive and negative effects.

Options for future researchA SID3 has been submitted detailing a proposal to continue investigations into culling-induced perturbation of the population in I2, consistent with that described in this report.

A proposal to quantify levels of neophobia and trap-shyness in a population subjected to culling has been submitted with the above SID3.

A more extensive study to assess the sustainability of culling by measuring some of the most important parameters related to social perturbation in I2 and other triplet areas of the RBCT would enable greater generalisation of results.

The perturbation study area provides a complementary low-density, disturbed population to that of Woodchester Park in with which to compare investigations of a range of potential management options to control M. bovis. Suggested priority areas of future research include the evaluation of novel methods to estimate population size, quantification of contact rates between badgers using proximity tags and studies of bait uptake.

An in depth analysis of the demographic and associated epidemiological data collected throughout the RBCT, in comparison with data collected during the perturbation study in I2, would be extremely useful to identify individual characteristics of culls in relation to their success.

Project Report to Defra

SID 5 (2/05) Page 4 of 31

8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

Main report – outline

1. Background

This report covers all activities carried out under Defra funded project SE3032 (1st April 2004 to 31st March 2007) that relate to the demographic, ecological and epidemiological consequences of culling badgers in the proactive triplet I2 of the RBCT. There is a compelling body of evidence supporting the hypothesis that social perturbation of badger populations following culling may be detrimental to the management of bovine TB in cattle. Field studies indicate that culling-induced social perturbation of badger populations includes a range of behaviours such as rapid immigration into culled areas, disruption of territoriality, increased ranging and increased movement of individuals between social groups (Cheeseman et al. 1993; Roper & Lüps, 1993; O’Corry-Crowe et al. 1993; Tuyttens et al. 2000a; Woodroffe et al. 2005a). Studies at Woodchester Park provided valuable information on the process of recolonisation following badger removal (Cheeseman et al. 1993) and field studies at North Nibley, Gloucestershire provided evidence of demographic disruption following culling (Tuyttens et al. 2000b). These studies showed the disturbance to social group territories and associated movements of individual badgers that followed, raising concerns that such social perturbation could potentially result in increased rates of disease transmission. Recent results from the RBCT show that cattle herd breakdowns can increase in some areas in response to badger culling (Donnelly et al. 2003; Le Fevre et al. 2005; Donnelly et al. 2006). A statistical correlation between rates of movement of badgers between social groups and the incidence of new M. bovis cases in the Woodchester Park population (Rogers et al. 1998) is consistent with the theory that social disruption may be linked to increased rates of disease transmission. Further analyses showed that social groups undergoing changes in size (whether increasing or decreasing) were associated with increases in the incidence of infection (Vicente et al. 2007), a phenomenon likely to be related to dispersal movements of badgers. Understanding the demographic and epidemiological consequences of badger culling is necessary in order to assess both the efficacy and sustainability of lethal control in reducing risks to cattle.

2. Scientific objectives as set out in the contract

The scientific objectives of the perturbation study were to conduct investigations into culling-induced perturbation of a badger population by:

1. Obtaining data on the spatial configuration of badger social groups subjected to culling (by bait marking).2. Collecting data on the size, structure and infection status of a population subjected to culling (by capture-

mark-recapture and clinical sampling).3. Collecting data on the behaviour of individual badgers in response to culling operations (by radio-

tracking).4. Obtaining data on the efficiency of badger culling operations (by comparison with trapping, bait marking

and radio-tracking data).

3. Extent to which objectives were met

All objectives have been met in full.

4. Methods

SID 5 (2/05) Page 5 of 31

4.1 Study area

The perturbation study area is situated within and adjacent to the proactive triplet I2 of the RBCT (centred on coordinates 2o26’W, 51o35’N) on the western edge of the Cotswolds. It covers 27.34 km2 and is subdivided into a 16.47 km2 area, which was culled between 2002 and 2005 (Table 1), and a 10.87 km2 area that has not been subjected to culling. The former area is referred to as the “culled area” throughout this report and the latter as the “unculled” area. The study area is relatively flat and composed predominantly of agricultural grassland and cereals.

Table 1 Schedule of fieldwork in the perturbation study area in relation toRBCT culling operations (in bold). Badgers found dead in the studyarea have been collected and sent for post mortem since June 2004 untilpresent.

Field technique/cull DateInitial cull2nd cullInitial signing up and sett surveyStart live-trapping, clinical sampling & radio-trackingCollection of faecal DNA3rd cullBait marking4th (final) cullBait markingFinish radio-trackingEnd live-trapping & clinical samplingBait marking

Sept/Oct 2002Sept/Oct 2003March-May 2004June 2004Sept-Nov 2004Oct/Nov 2004Feb-April 2005July 2005Feb-April 2006Dec 2006Jan 2007Feb-April 2007

4.2 Bait marking (Objective 1)

The configuration of badger social groups was determined by bait marking (see Delahay et al. 2000a). Bait marking provides an estimate of the home range of a social group; it can identify annual changes in the size of group ranges and the degree of range exclusivity indicates the level of territoriality in the population. A palatable bait of peanuts and syrup laced with small coloured plastic pellets was fed for a minimum of 12 days at each active sett. The subsequent distribution of the harmless indigestible pellets in faeces at badger latrines was used to map social group territories using a geographic information system (ArcView 3.2 and ArcGIS 9.1: Environmental Systems Research Institute (ESRI), Redlands, California, USA, 2005). Bait marking was conducted each spring throughout the entire perturbation study area between 2005 and 2007. This study began in April 2004, consequently it was not possible to conduct a spring bait marking survey in that year. A 2007 bait marking survey was completed in April 2007 to maximise information on the spatial organisation of this population, the results of this survey are included in this report.

4.3 Live-trapping (Objectives 2 & 4)

Demographic and epidemiological information on the resident badger population was collected by means of a routine capture-mark-release (CMR) programme, consistent with that carried out at Woodchester Park since 1981 (see Rogers et al. 1997). Trapping took place throughout the year, except for during a ‘close season’ in the spring (February to April inclusive) when female badgers may have small dependent cubs that cannot be left alone for protracted periods (see Woodroffe et al. 2005b). Badger social groups in the study area were trapped on average four times a year between 2004 and 2006. The entire area was surveyed in the spring of 2004 to identify all active and inactive setts in the area. Setts (active and inactive) were subsequently visited prior to each trapping operation to determine their level of activity. Traps were deployed by digging them into the substrate in the immediate vicinity of each active sett. Levels of activity and knowledge of the number of animals previously caught were used to guide the number of traps per sett, although the general approach was to put out more traps than were likely to be needed (i.e. saturation trapping). Remote trapping, whereby traps were placed on runs away from setts, was also conducted in the culled area in 2004 in an attempt to increase the number of radio-collared badgers in the culled area.

Traps were of steel mesh box design, with a spring-loaded door mechanism. Once traps had been deployed they were secured in the open position and baited daily with peanuts for four to eight days to habituate the badgers to their presence (Cheeseman & Mallinson 1979). On the last day of this pre-baiting period the traps were set for two consecutive nights and checked on each following morning. From December until the ‘close season’ began (February 1st), the traps were also checked at night so that small cubs and lactating females could

SID 5 (2/05) Page 6 of 31

be released (in the case of adult animals an attempt was made to read and record their tattoo prior to release). After capture, each badger was transferred to a holding cage before being transported back to the sampling facility. In order to reduce opportunities for the transmission of disease between badgers from different social groups, they were segregated during transport in vehicles and arranged into their respective groups on racks in the sampling facility. In addition, holding cages were not exchanged between social groups during trapping operations and were thoroughly cleaned between operations.

Animals were trapped and examined for two consecutive days. Those caught during the first night of trapping were held overnight after their examination and sampling, and normally released the following morning. This prevented them from being re-captured on the second trapping night. However, lactating females or small cubs were marked (with an aerosol stock marker) and released on the same day. If these marked animals were recaptured they were immediately released. Animals caught and examined on the second day were released later that day. Badgers were released at the point of capture, following a period of recovery and subject to approval by the named Animal Care and Welfare Officer (NACWO), Named Veterinary Surgeon (NVS) or another experienced person with delegated authority.

4.4 Clinical sampling (Objective 2)

Techniques developed at Woodchester Park to investigate population structure and TB epidemiology (see SE3032) have been employed to study the population in I2 and have also been used in earlier studies of perturbation (e.g. Tuyttens et al. 2000b). Although imperfect, this methodology is the only means currently available to monitor disease dynamics in live wild badgers. Trapped badgers were anaesthetised with a mixture of ketamine hydrochloride, medetomidine hydrochloride and butorphanol tartrate (de Leeuw et al. 2004), and on first capture each was given a unique identifying tattoo (Cheeseman & Harris 1982) and aged (cub vs adult) according to characteristics such as body size, tooth development and tooth wear. For the purposes of determining TB infection status clinical samples (for microbiological culture) and blood (for serology) were taken. A blood sample was collected from the jugular vein for the detection of antibodies to the single M. bovis antigen MPB83 using an ELISA test (Goodger et al. 1994). All blood samples were processed by the Veterinary Laboratories Agency (VLA), Langford, Somerset. Clinical samples of sputum were obtained by oesophageal aspiration until May 2006, after which they were collected by tracheal aspiration. This change was made on veterinary advice, which suggested that tracheal sampling may be more sensitive than oesophageal sampling (owing to lower levels of contamination in the former). Urine was obtained by palpation of the bladder, faeces by administration of an enema and swabs of pus and exudate were taken from bites and open abscesses. All M. bovis isolates were sent to VLA, Weybridge for spoligotyping (Aranaz et al. 1996).

4.5 Post mortem examinations (Objective 2)

All badgers found dead within the study area were submitted to the VLA Regional Laboratory at Luddington, Warwickshire for post mortem examination. Selected tissues were submitted to CSL’s Microbiology Group at York for culture and isolates forwarded to CVL at Weybridge for spoligotyping.

4.6 Radio-tracking (Objective 3)

All adult badgers captured in the study area between June 2004 and November 2006 were fitted with radio-collars to determine their home ranges and sett use in relation to culling operations. Radio-collars consisted of TW-3 transmitters with a closed loop antenna (Biotrack Ltd, Furzebrook, Wareham, Dorset, UK), encased in epoxy resin and set into a leather collar. Collars were not fitted to badgers with severe lesions or wounds to their neck. Collar weight was well below 5% of a badger’s body weight as recommended for radio-tracking studies (Cochran 1980). Radio tracking was carried out by car and on foot using a hand-held Yagi-flexible-element antenna (Biotrack Ltd, UK) connected via a coaxial cable to a TR-4 receiver (Telonics Inc., 932E Impala Avenue, Mesa, AZ, USA). Landmarks were used wherever possible to map the position of the observer and/or the badger being tracked. Precise locations of specific vantage points from which a clear signal was known to be detectable were recorded and digitally mapped using ArcGIS 9.1 (ESRI, 2005). Bearings were taken using a mirror compass and observer location was recorded with a GPS (DETAILS?) when landmarks were absent. Nocturnal ranging data were collected mainly between the hours of 21:00 and 05:00, adjusted according to season and subsequent emergent patterns. Daytime location data for each collared badger was recorded by visiting all known setts in the study area once a week for the duration of the study. Individual home ranges for each badger were analysed using Ranges 6, Version 1.2 (Anatrack Ltd., 2003).

4.7 Genotyping of faecal DNA

In light of anecdotal evidence of trap-shyness among badgers in the culled area (see 4.7.3 below) we conducted a complementary piece of research (not part of the objectives set out in the contract) in collaboration with Sheffield University that attempted to provide an estimate of population size by genotyping DNA recovered from

SID 5 (2/05) Page 7 of 31

badger faeces collected from latrines throughout the perturbation study area prior to the 2004 cull. Whilst this technique has enormous potential for estimating badger density and recolonisation rates, the pilot study revealed a need for further optimisation of methodology before it can be applied in the field. The full results of this pilot study can be found in the report from Sheffield University (Appendix I).

4.8 Data analyses

4.8.1 Social organisation

Social group home ranges were defined as the area included in a 100% minimum convex polygon (MCP) around the outermost bait returns (latrines/faecal deposits containing plastic pellets), after excluding outlying bait returns and adjusting the MCP to accommodate boundary runs as per Delahay et al. (2000). This method may underestimate group range size where there are few bait returns, such as following culling. We therefore calculated the mean distance between a sett and its associated bait returns as an alternative measure in accordance with Tuyttens et al. (2000a) and Woodroffe et al. (2005). Group range size and overlap were determined using ArcGIS 9.1 (ESRI 2005). We measured the proportion of each group home range that overlapped with other group home ranges and we recorded the number of shared latrines per social group as a measure of exclusivity and contiguity. The number of bait returns per sett, social group area, and bait return distance were compared between culled and unculled areas, and between years using linear mixed models (GenStat Version 9.1: Lawes Agricultural Trust, 2006). Treatment (culled, unculled) and year were fitted as fixed factors, whilst sett (for analysis of number of bait returns and bait return distance) or social group (for social group area analysis) were fitted as random factors. Data not assuming a normal distribution were log-transformed before analysis. A single return recorded in the culled area a little over 2 km from the sett at which it was fed explained a disproportionate amount of the variance in mean group range size in 2005. There were no additional ecological data to disregard this return in the calculation of the social group home range (See Delahay et al. 2000). However, because it was substantially further than the next recorded distance in that year (1.2 km) the analysis was repeated treating this return as an outlier.

4.8.2 Demographic analyses

The minimum number alive (MNA) method (Rogers et al. 1997) was used to estimate the population size each year and, as the size of the study area did not change, as a proxy for badger density. There were not sufficient recapture data to calculate reliable estimates of annual trap and survival probability using a Jolly-Seber mark-recapture analysis for the different groups and years. Recapture rates represent the percentage of captured animals recaptured in the following year (minus known fatalities), although the failure to re-trap animals will be a function of both capture probability and mortality, the relative importance of which cannot in this case be reliably estimated from the data. In addition to trapping records, information from animals radio-tracked and captured by the Wildlife Unit during culling was also included. A long-term capture mark recapture study at Woodchester Park has demonstrated that the majority (89%) of animals captured are caught as cubs in their year of birth. The number of unmarked adult badgers captured or found dead may consequently be used as a crude proxy for immigration. Previous studies of perturbation have employed this approach in an attempt to measure recolonisation. The number and sex of unmarked adults captured or found dead in the perturbation study area in 2005 and 2006 was recorded for comparison with similar data following a badger removal operation at North Woodchester (Cheeseman et al. 1988). Statistical difference between the ratio of cubs to adults in culled and unculled areas was assessed using chi-squared tests based on absolute numbers of cubs and adults in each area derived using the MNA method. Separate tests were carried out for each year (2004, 2005 and 2006), as the data from successive years were non-independent.

4.8.3 Analysis of trap-shyness

During the initial live-trapping operation it was observed that a number of setts that had initially shown signs of badger activity at the time they were first surveyed and traps were dug into position, had then become inactive. It was hypothesised that human disturbance to the setts may have prompted badgers to move to different setts and that previous disturbance in the culled area may have resulted in a neophobic population more prone to disturbance. Subsequently it was recorded whether each sett that was active during the pre-trapping survey was still active following pre-baiting, prior to setting traps to catch badgers. Setts were deemed to be active if there was evidence of any of the following: bait being taken, fresh bedding, fresh spoil or recent footprints/claw-marks. The proportion of setts remaining active was compared between treatment areas. As badgers may exhibit neophobia in the form of “trap-shyness” i.e. a reluctance to enter traps, without actually moving setts it was recorded whether badgers were caught or not from setts deemed to be still active at each trapping occasion. The probability that setts were still active, and the probability that badgers were trapped at a given sett on a given occasion were modelled after logit-transformation of the binary responses (1 = active or badgers trapped; 0 = inactive or no badgers trapped) using generalised linear mixed models. Treatment area, a

SID 5 (2/05) Page 8 of 31

linear effect of time (measured by trapping occasion fitted as a continuous variable) and season (4-level categorical, Spring Mar-May, Summer Jun-Aug, Autumn Sep-Nov and Winter Dec-Feb were included as fixed effects and assessed with Wald tests. Sett and all observations from the same sett were included as random effects. Each model was run again including an interaction between treatment and time (trap occasion) to detect any difference in trends between the culled and unculled areas.

4.8.4 Home range analysis

Home ranges were analysed using Kernel analysis (Kenward et al, 2003), whereby the proximity of locations to each evaluation point are summed in order to obtain a density estimate on which contours are modelled. Kernels are not as sensitive to outliers as MCPs, and less sensitive to the problems of autocorrelation than parametric methods (Seaman et al. 1998). Contours were based on fix density and estimated the probability of including a percentage of fixes. Only fixes recorded whilst badgers were active above ground were used in the analysis of ranging patterns (Seaman et al. 1998). Utilisation plots (Kenward et al. 2003) were used to determine the percentage of fixes that constituted the home range and core areas by displaying the area of estimated home range cores at 5% intervals from 20-100%. Examination of utilisation plots revealed that an average of 92% fixes gave reliable home range estimates with excursive movements excluded from the range area, while an average of 52% fixes gave stable core area estimates, using a reference smoothing parameter1 with a multiplier of 0.6. Incremental area analysis (Kenward et al. 2003) was used to determine that sufficient radio-fixes were obtained to define the home range of individual badgers that were radio-tracked during the study period. Consecutive areas were plotted against the number of locations (fixes) for each badger until an asymptote was reached as evidence of stability (this occurred between 40 and 80 fixes) indicating that further locations would not improve the home range estimate. Badger ranges with an insufficient number of fixes to yield a stable home range estimate (no asymptote was reached) were excluded from subsequent analysis. To examine the effects of season or successive culling operations it was necessary to truncate periods of tracking on individuals. Incremental area analysis was used on badger ranges for a single interval (post-cull 2, June 2004 to October 2004) to determine the average number of fixes necessary to estimate home ranges of badgers during similar truncated periods of tracking. This indicated that 40 ± 2.0 fixes were required, so all ranges with fewer than 38 fixes were, conservatively, excluded from the analysis of the effects of season and successive culling. Linear mixed models (Genstat 9.1, 2006) were used to analyse the effects of treatment (culled, unculled) and year on home range area. Treatment, sex and log number of fixes per range (to account for variation in the number of fixes obtained per individual) were fitted as fixed factors, with badger ID and sett (as a proxy for social group) fitted as random factors. Season (defined above) and culling interval (post-cull 2, June 2004 to October 2004; post-cull 3, November 2004 to June 2005; post-cull 4, August 2005 to Nov 2006) were fitted as additional fixed factors to models examining the effect of either.

4.8.5 Analysis of prevalence

Differences in the prevalence of M. bovis infection in badgers between the two treatment areas was investigated by fitting generalised linear mixed models to the results of culture and ELISA testing respectively. The results of the collective culture results for a given trap event were summarised in binary form: 0 = no positive results; 1 = one or more positive results. This binary response was logit-transformed and the probability of a positive result modelled as a function of the following putative explanatory variables: year (3-level categorical, i.e. 2004 through to 2006), age (cub/adult), sex and treatment (culled, unculled). Badger ID and the sett where each animal was trapped were incorporated as random effects because of anticipated pseudo-replication in results from the same source. Statistical inference on the above explanatory variables was based on Wald tests. The probability of a positive ELISA test was modelled with the same random and fixed effects fitted.

5. Results

5.1 The effect of culling on badger social organisation

A total of 29 (culled = 15; unculled = 14) setts were fed marked bait between 2005 and 2007. The number of setts fed each spring varied according to sett activity levels and landowner permission: n = 17, 2005; n = 25, 2006; n = 23, 2007 (Fig. 1). Bait marking indicated up to 22 discrete social groups in the study area (culled = 11; unculled = 11); this number varied between 14 and 20 per year during the three years of the study. The mean number of latrines containing plastic markers (bait returns) per sett over the three years of the study varied from 6.33 to 8.83 in the culled area and 7.3 to 11.63 in the unculled area. There were twice as many returns per sett in the unculled area vs the culled area in 2005, but the mean number of returns per sett was not correlated with either year or treatment (2 = 0.2, df = 1, P = 0.65; 2 = 1.03, df = 2, P = 0.6, respectively). The number of returns per social group was correlated with social group area in 2005 and 2006 (RS= 0.79, df = 12, P < 0.01; RS = 0.86, df = 18, P <

1 The standard deviation of x and y coordinates divided by 6th root of sample size.

SID 5 (2/05) Page 9 of 31

0.01, respectively), and it approached significance in 2007 (RS = 0.49, df = 14, P < 0.06). There was no such correlation between the number of returns per sett and the distance from setts to their associated returns in any year (2005, RS = 0.16, df = 15, P = 0.54; 2006, RS = 0.25, df = 22, P = 0.25; 2007, RS = 0.11, df = 18, P = 0.65). This suggests that the latter may be a more informative measure of badger ranging behaviour (determined from bait marking) in accordance with Woodroffe et al. (2005). Analyses have been conducted on both measures for consistency and comparability with other studies.

SID 5 (2/05) Page 10 of 31

Figure 1 Spatial configuration of badger social group home ranges in the perturbation study area in 2005, 2006 and 2007. Stars show the location of bait marked setts, black solid lines denote minimum convex polygons around bait returns, the grey solid line represents the boundary of the study area and the dotted line shows the boundary of the proactive culling triplet I2.

SID 5 (2/05) Page 11 of 31

unculled culled

unculled culled

unculled culled

5.1.1 Differences in social group home range area

The mean social group range was almost three times larger in the culled area in 2005 than in the unculled area (Fig. 2) including the outlying point (see above) and almost twice that in the unculled area excluding this point. In addition, the mean social group range in the culled area appeared to decrease dramatically after 2005. Nethertheless, there was no significant effect of treatment or year regardless of whether the outlier was included (treatment: 2 = 0.64, df = 1, P = 0.43; year: 2 = 0.80, df = 2, P = 0.67) or not (treatment: 2 = 0.20, df = 1, P = 0.65; year: 2 = 0.70, df = 2, P = 0.7). The interaction between treatment and year was also not significant (incl. outlier: 2 = 3.41, df = 2, P = 0.18; excl. outlier: 2 = 1.39, df = 2, P = 0.5). Despite all nine of the active setts in the culled area being fed marked bait in 2005, uptake was considered to be low for about half and a quantity of bait returns sufficient to calculate group home ranges were only found for seven setts, which equated to six discrete social groups. In two cases social group areas were estimated from only three returns. Very few social groups overlapped in either culled or unculled areas for the duration of the study and the overlapping areas generally constituted a small percentage of the social group range (Table 2). Whilst the group ranges in the unculled area appeared more contiguous (Fig. 1), an index of contiguity (average of the number of shared latrines divided by the total number of latrines per social group) ranged between 6.4 and 22.6 in the unculled area and 13.1 and 22.5 in the culled area with no obvious annual pattern.

Figure 2 Mean social group home ranges (± 1 SE) in the culled and unculled areas of the perturbation study area for the three years of the study in relation to RBCT culling operations.

Table 2 Mean percentage area of social group home ranges overlapping with other home ranges during the three years of the study.Mean % overlap 2005 2006 2007unculled groups 28.09 (n = 1) 1.32 (n = 6) 5.39 (n = 4)culled groups 0 22.89 (n = 1) 0unculled-culled groups 0 2.22 (n = 1) 0

5.1.2 Differences in bait return distances

The interaction between year and treatment was highly significant (2 = 9.69, df = 2, P < 0.01), such that bait returns were on average 140 m (45%) further from setts in the culled area in 2005 and 2007, but approximately 25 m closer in 2006 (Fig. 3). There was no effect of treatment (2 = 1.27, df = 1, P = 0.26) on the mean distance

SID 5 (2/05) Page 12 of 31

0

20

40

60

80

100

120

140

2005 2006 2007

Mea

n so

cial

gro

up a

rea

(Ha)

unculled culled

Cull 4 Cull 3

between setts and their associated bait returns, but there was a significant effect of year (2 = 6.7, df = 2, P = 0.04).

Figure 3 Mean distance between setts and their associated bait returns (± 1 SE) in the culled and unculled areas of the perturbation study area for the three years of the study in relation to RBCT culling operations.

5.2 Demographic consequences of culling

Over the course of the three years of the study there were 244 capture events during a total of 2,646 trap nights, accounting for 134 individual badgers. These were captured from 18 of the 22 social groups identified by bait marking. There was no landowner permission to trap at a sett within one social group during the three years of the study, although permission to bait mark was granted in 2006. Permission to trap at another sett within a different social group was periodically withdrawn, but the sett was bait marked each year. Both setts occured in the unculled area. Annual recapture rates for the whole study area were considerably higher in 2005 than 2006 (36% vs 15%). This was largely the result of higher recapture rates in the unculled area in 2005 (39% vs 6%). Recapture rates in the culled area were lower than in the unculled area in 2005 (22% vs 36%), but considerably higher than in the unculled area in 2006 (30% vs 6%). Mean social group size was slightly lower in the culled area and more variable over the three years of the study (unculled, mean = 4, SE = 0.19; culled, mean = 3.1, SE = 0.72). Demographic data were also available for six unmarked individuals found dead in the study area and an additional 16 unmarked animals culled or found dead by the Defra Wildlife Unit in 2004 and 2005. The estimated number of badgers in the population each year (derived using the MNA method) fluctuated between 57 and 76 (Fig. 4). The number of adults fluctuated between 20 and 38 with no obvious annual pattern, but there were consistently about twice as many adults in the unculled area than in the culled area. In comparison the number of cubs remained fairly constant between years with no obvious differences between the culled and unculled areas (Fig. 4). The MNA method suggested a population density of 3.65 (SE 0.21) badgers km -2 in the unculled area and 1.74 (SE 0.21) badgers km-2 in the culled area with very little variation between years.

SID 5 (2/05) Page 13 of 31

0

100

200

300

400

500

600

2005 2006 2007

Bai

t ret

urn

dist

ance

(m)

unculled culled

Cull 3 Cull 4

Figure 4 Estimated numbers of resident cubs and adult badgers (derived using the MNA method) in the culled and unculled regions of the perturbation study area (2004-2006) based on live-trapping records, radio-tracked badgers, RTAs and culled badgers. The latter relates to 2004 and 2005 only.

The ratio of cubs to adults between culled and unculled areas was equal in 2004, but was significantly higher in the culled area in 2005 (2 = 5.3, df = 1, P = 0.02) and 2006 (2 = 4.8, df = 1, P = 0.03), indicating that the population in the culled area was becoming younger relative to the adjacent unculled population (Fig. 5).

Figure 5 Ratio of cubs to adults in the culled and unculled regions of the perturbation study area (2004-2006). Equivalent data for an undisturbed populationat Woodchester Park are shown for comparison. The ratios are derived from theestimated total number of badgers each year using the MNA method. Culls 2-4 (Table 1) are schematic indicating the sequence of culling relative to the timing ofbirths each year.

The ratio of adult females to adult males was five times higher in the culled area than in the unculled area in 2005 (2 = 3.8, df = 1, P = 0.051) and more than twice as high in 2006 (2 = 3.3, df = 1, P = 0.07) indicating that the population had become more female biased in the culled area after 2004.

SID 5 (2/05) Page 14 of 31

01020304050

unculled culled unculled culled unculled culled

2004 2005 2006

Num

ber o

f bad

gers

(MN

A) cubs adults

0

0.5

1

1.5

2

2.5

3

2004 2005 2006

Cub

:adu

lt ra

tio

unculled culled WP

Cull 4 Cull 3 Cull 2

Figure 6 Ratio of adult females to adult males in the perturbation study area Between 2004 and 2006. The ratios are derived from the estimated total number of adult badgers each year using the MNA method. The broken line represents a 1:1 sex ratio.

5.2.1 Evidence of trap-shyness

A significantly higher proportion of setts in the unculled area to remained active throughout the period of pre-baiting before traps were set to catch badgers (Table 3a; Fig. 7), with a tendency for the difference between treatments to decrease over time. A higher proportion of trapping operations at active setts in the unculled area caught badgers (Table 3b; Fig. 8). Season had a significant effect in both models indicating that setts were more likely to remain active and to yield badger captures in spring and least likely in winter.

Table 3 Coefficients and Wald tests from a GLMM to partition variation associated with (a) the probability of observed setts remaining “active” during pre-baiting, and (b) the probability oftrapping 1+ badger(s) at an active sett.

(a)Model output

Variable Categorical levels

-coefficient Wald (df) P-value

treatment 11.3 (1) < 0.001unculled 0culled - 4.3

season 13.7 (3) 0.003spring 0

summer - 0.7autumn - 1.9winter - 2.0

time + 0.5 6.6 (1) 0.001

(b)Model output

Variable Categorical levels

-coefficient Wald (df) P-value

treatment 5.7 (1) 0.02unculled 0culled - 1.2

season 33.6 (3) < 0.001spring 0

summer - 0.2

SID 5 (2/05) Page 15 of 31

0

1

2

3

4

5

6

2004 2005 2006

Fem

ale:

mal

e ra

tiounculled culled

autumn - 1.2winter - 2.7

time + 0.1 1.6 (1) 0.2

Figure 7 The proportion of setts in the perturbation study area that remained active throughout each pre-baiting period, and that were still active when trapping commenced. The arrows indicate the timing of the 3rd (2004) and 4th (2005) culling operations. The initial and 2nd culls occurred in October of 2002 and 2003 respectively (Table 1).

Figure 8 The proportion of active setts in the culled and unculled portions of the perturbation studyarea from which badgers were caught during each trapping operation. The arrows indicate the timing of the 3rd (2004) and 4th (2005) culling operations. The initial and 2nd culls occurred in October of 2002 and 2003 respectively (Table 1).

SID 5 (2/05) Page 16 of 31

0

0.2

0.4

0.6

0.8

1

Jun04

Aug04

Dec04

Jan05

May05

Sep05

Nov05

Jan06

Jun06

Aug06

Nov06

Jan07

Live-trapping operation

Prop

ortio

n of

set

ts s

till a

ctiv

e

Culled Unculled

0 0 0 00

0.2

0.4

0.6

0.8

1

Jun04

Aug04

Dec04

Jan05

May05

Sep05

Nov05

Jan06

Jun06

Aug06

Nov06

Jan07

Live-trapping operation

Prop

ortio

n of

act

ive

setts

whe

re

badg

ers

wer

e ca

ught

Culled Unculled

5.2.2 Culling efficiency

Figure 9 shows the demographic composition of the badger population in the culled portion of the perturbation study area (based on the MNA approach) between 2004 and 2006. Figure 10 shows the demographic composition of badgers culled from this area between 2002 and 2005. The data displayed in Fig. 10 were included in the calculation of the MNA population estimate. Consequently, the similar pattern observed for the years in which they overlap is not unexpected. Crude culling efficiency estimates [(number removed/MNA population estimate)*100] for 2004 and 2005 are 50% and 62%, respectively. There was considerable age and sex-related variation in culling efficiency estimates within and between years, but they tended to be higher for females than for males (Figs. 9 & 10). Approximately 10% to 13% of badgers are believed to be trap-shy (C. L. Cheeseman unpublished observations), consequently the MNA method is likely to underestimate population size by about this percentage when live-trapping has predominantly been used to estimate population size, as in this study. However, the population estimate was not increased to account for trap-shyness as a similar percentage of the annual population estimate would not have necessarily been in the population at the time a cull took place e.g. animals live-trapped prior to culling and never re-trapped but whose fate was unknown. The comparison of unique genotypes from faecal material collected from latrines in the culled area prior to the 2004 cull, with genotypes obtained from culled individuals suggested that approximately 22% (2 of 9 faecal genotypes present in cull) to 54% (6 of 11 culled individuals matched faecal genotypes) of individuals for which faecal genotypes were obtained were culled.

Figure 9 Sex and age composition of the badger population in the culled regionof the perturbation study area between 2004 and 2006, derived from the MNA method.

SID 5 (2/05) Page 17 of 31

0

5

10

15

20

25

30

Males Females Males Females Males Females

2004 2005 2006

Num

ber o

f bad

gers

(MN

A)

cubs adults

Figure 10 Age and sex composition of badgers culled from the perturbation studyarea between 2002 and 2004.

5.2.3 Dispersal

Out of a total of 244 capture events and 1,356 daytime positions, eight movements between social groups were recorded. These movements involved five badgers, one in the culled area and four in the unculled area. No movements of badgers were recorded between the areas. All movements were by adult males, except for one female cub. Two badgers, one in the culled area and one in the unculled area, were detected to have moved using daytime location data. In each case they were located in a sett belonging to a neighbouring group on one occasion and then located within their original social group on a subsequent occasion. The remainder involved three badgers that were trapped at setts within different social groups in the unculled area. In two cases each badger was trapped only twice, initially in 2004 and once more within a neighbouring social group in 2005 and 2006. The female cub was trapped three times in 2006, the second time at a sett belonging to a neighbouring group and the third within the original group. There were no movements (defined as the use of a sett belonging to a different social group) between social groups in culled and unculled areas detected from radio-tracking.

The number of unmarked adults captured or found dead in the study area in 2005 and 2006 was generally low in both culled (2005, n = 2; 2006, n = 4) and unculled areas (2005, n = 6; 2006, n = 3), but there were more unmarked females (4) than males (2) in the culled area in both years combined.

5.3 Consequences of culling on individual behaviour and movement patterns

A total of 40 badgers were fitted with radio-collars between 2004 and 2006. This accounted for all adult badgers caught in the study area, other than those with severe lesions or wounds to the neck preventing the fitting of a collar (n = 8). Thirteen radio-collared badgers were captured in the culled area and 27 in the adjacent unculled area. A total of 2,946 individual radio fixes were collected during 1,248 hours of nocturnal tracking and 1,356 daytime locations. Incremental area analysis indicated that there were sufficient data to generate accurate home range estimates for 30 badgers, 9 in the culled area and 21 in the unculled area, the remainder being lost from the study due to a combination of fatalities, premature shedding of collars and either long distance dispersal or collar malfunction.

5.3.1 Sex differences in home range size

There were significant differences in core range size (52% kernels) between the sexes although home range size (92% kernels) did not differ. Male core areas were approximately 50% larger than those of females (52% kernels; t = 2.96, P < 0.01), whilst male home ranges were around 30% larger (92% kernels; t = 1.79, P = 0.09). There was also an imbalance of adult male and female badgers captured in each area such that after excluding animals with insufficient fixes to determine home ranges with any confidence (see above) data were available for four males and five females in the culled area and 13 males and eight females in the unculled area. Sex was included in all following models as a fixed factor to account for any sex related bias.

SID 5 (2/05) Page 18 of 31

0

4

8

12

16

20

24

Mal

es

Fem

ales

Mal

es

Fem

ales

Mal

es

Fem

ales

Mal

es

Fem

ales

2002 2003 2004 2005

Bad

gers

cul

led

Cubs Adults

5.3.2 Differences in individual home range size and overlap

Over the whole study, the core ranging areas of individuals in the culled area were approximately twice as large as those in the unculled area (2 = 6.19, df = 1, P = 0.013; Fig. 11). The size of core areas was also found to vary significantly with respect to sex (2 = 14.09, df = 1, P < 0.001) and the log number of fixes (2 = 4.41, df = 1, P = 0.036). Home range areas were 60% larger in the culled area (Fig. 11), but varied significantly with respect to sex (2 = 5.91, df = 1, P = 0.015) and log number of fixes (2 = 5.86, df = 1, P = 0.015), such that the variance between culled and unculled areas only approached statistical significance (2 = 3.05, df = 1, P = 0.081). No overlapping home ranges were recorded for badgers from different social groups in the culled area, consequently differences in home range overlap between culled and unculled groups could not be analysed.

Fig. 11 Mean (± 1 SE) kernel home range (92%) and core area (52%) estimatesfor badgers in the perturbation study area over the three-year study period (Datacollected between June 2004 and November 2006).

5.3.3 Differences in home range size with successive culls

Between the start of radio-tracking in June 2004 and the end of tracking in November 2005 home ranges in the culled area were significantly larger than those in the unculled area between the third and fourth follow-up cull, and following the fourth cull (92% range: 2 = 4.42, df = 1, P = 0.036; Table 4; Fig. 12). Home ranges in the culled area also showed a tendency to increase in size in the culled area with successive culling operations (significant interaction between treatment (culled, unculled) and interval between culling: 2 = 344.08, df = 2, P < 0.001). None of the remaining terms in the model were significant (interval between culling: 2 = 1.98, df = 2, P = 0.37; sex: 2 = 3.25, df = 1, P = 0.072; log number of fixes: 2 = 0.02, df = 1, P = 0.876). Core areas of badger ranges were also larger in the culled area (52% core: 2 = 7.37, df = 1, P = 0.007) and followed the same pattern as the 92% home ranges, with a tendency to increase in size in the culled area with successive culling operations (Table 5) with virtually the opposite trend in the unculled area (interval between culling, 2 = 13.31, df = 2, P = 0.001). Sizes of male and female core areas varied significantly between culling operations (sex: 2 = 7.18, df = 1, P = 0.007), but there was no significant variation in the number of fixes used to delineate core areas (log number of fixes: 2 = 1.08, df = 1, P = 0.3).

Table 4 Home range (92% kernel) sizes of badgers tracked during different time intervals between and following culling operations in I2. Post-cull 2 represents the period betweenthe start of this study and the 3rd culling operation (June – October 2004), post-cull 3represents the period between the end of the 3rd culling operation and the 4th culling

SID 5 (2/05) Page 19 of 31

0

10

20

30

40

50

60

70

Core area (52%) Home range (92%)

Ker

nel r

ange

are

as (H

a)

culled unculled

operation (November 2004 – July 2005) and post-cull 4 represents the period betweenthe end of the 4th culling operation and the end of the collection of nocturnal radio-trackingdata for this study (August 2005 – November 2006).

Period Culled area Unculled area Combined areaMean (Ha)

SE Mean (Ha)

SE Mean (Ha)

SE

Post-cull 2Post-cull 3Post-cull 4Total period

34.6847.9661.5951.74

6.277.989.045.72

37.2522.7132.9430.40

8.016.118.134.24

36.7829.4543.1736.27

6.555.677.043.74

Table 5 Core areas (52% kernel) of badgers tracked during different time intervalsbetween and following culling operations in I2. Post-cull 2 represents the period betweenthe start of this study and the 3rd culling operation (June – October 2004), post-cull 3represents the period between the end of the 3rd culling operation and the 4th cullingoperation (November 2004 – July 2005) and post-cull 4 represents the period betweenthe end of the 4th culling operation and the end of the collection of nocturnal radio-trackingdata for this study (August 2005 – November 2006).

Period Culled area Unculled area Combined areaMean (Ha)

SE Mean (Ha)

SE Mean (Ha)

SE

Post-cull 2Post-cull 3Post-cull 4Total period

10.5613.2421.7916.64

0.361.894.982.68

11.506.688.998.89

2.481.523.031.35

11.338.4313.5611.02

2.011.423.041.33

Figure 12 Differences in mean (± 1 SE) home range sizes for badgers in the culledand unculled areas following the three follow-up culls in the perturbation study area.

5.3.4 Seasonal differences in home range sizes

Home ranges varied according to the season that badgers were tracked although this effect only approached statistical significance (spring, summer and autumn included in model only: 2 = 5.88, df = 2, P = 0.053), but the size of core ranges did not (season: 2 = 4.53, df = 2, P = 0.104). In this model home range size and core ranging

SID 5 (2/05) Page 20 of 31

010203040

50607080

post-cull 2 post-cull 3 post-cull 4

92%

ker

nel h

ome

rang

e

culled unculled

areas did not vary significantly according to sex or the number of fixes (92% home range, sex: 2 = 0.03, df = 1, P = 0.859; log number of fixes: 2 = 3.01, df = 1, P = 0.083; 52% core, sex: 2 = 0.04, df = 1, P = 0.842; log number of fixes: 2 = 1.16, df = 1, P = 0.281).

5.4 Prevalence and distribution of infection

The repeated sampling of individual animals captured in the perturbation study area in the past three years has produced a database which currently contains records of the life histories of 140 badgers from 244 capture events and 10 records of dead animals found in the study area and submitted for post mortem. Annual estimates of prevalence (number of M. bovis culture positive individuals/number of individuals live-sampled per year) ranged between 6.4% and 11.5% for the entire study area (Table 6). No culture positive results were obtained from 58 individuals captured in the culled area during the three years of the study from 91 live-sampling events i.e. including recaptures of individuals (Table 6). Annual estimates of prevalence from live-sampling for the unculled area were 15.2% in 2004, 17.6% in 2005 and 11.5% in 2006, with an apparent increase in their spatial distribution in 2005 (Fig. 13). From 16 positive M. bovis culture samples collected over three years from badgers in the unculled area, seven were from sputum, two were from urine, one from faeces, and six from open wounds and abscesses. Two of these (one urine and one faecal sample) were collected from the same individual on the same occasion. There were no detectable differences in the probability of an ELISA positive result between badgers caught in the culled and unculled areas, however the interaction between treatment and year was significant (2 = 14.2, df = 2, P < 0.001; Table 7) indicating that there was no consistent pattern between the two areas over the three years of the study. Examination of the distribution of ELISA positive results between the culled and unculled areas revealed a substantial decrease in prevalence in 2005 in the culled area (unculled: 2004, 36%; 2005, 58%; 2006, 58%; culled: 2004, 39%; 2005, 5%; 2006, 26%; Fig. 13). Adults had a higher probability of a positive result from culture and ELISA than did cubs, although this was only significant for culture results (Tables 6 & 7). None of the badgers found dead in the study area cultured positive for M. bovis at post mortem.

The proportion of culture positive cases identified following post mortem examination and tissue culture of badgers trapped in the culled area during 2004 and 2005 showed that four out of eleven (36.4%) badgers culled in 2004 were found to be culture positive for M. bovis at post mortem and another three animals had visible lesions (but were not recorded as culture positive). Six of these were marked animals that had been captured as cubs earlier in the year and were found to be culture negative from live-sampling, although two out of three cubs that were live-sampled within three months of being killed were identified as being ELISA positive during live-sampling. The interval between the most recent live-sampling event and being killed during culling varied between seven and 19 weeks (mean 13.25, SD 5.8). In 2005 two out of 19 culled badgers were culture positive at post mortem; one of these was marked and had produced a negative result from live-sampling in the previous September but had not been caught again until it was trapped by the Defra Wildlife Unit in July 2005. Upon examination of the full post mortem results only one of the seven badgers previously live-sampled had profuse respiratory lymph node lesions. The remainder appeared to be in relatively early stages of the disease or the result was unclear. Annual estimates of prevalence (culture positives only) determined from post mortem examination of badgers culled in the perturbation study area varied considerably: 2002, 49% (18/37); 2003, 17.9% (5/28); 2004, 36.4% (4/11); 2005, 10.5% (2/19) (Fig. 14).

Table 6 Results of a generalised linear mixed model (GLMM) predicting the probability of a positive M. bovis culture result from live-sampling. Treatment (culled vs unculled), year, age and sex were fitted as main effects. Badger ID and sett were fitted as random effects.

Raw data Model outputvariable level n/N (%) culture

+ve trap eventsn/N (%) culture +ve individuals (per yr)

beta Wald (df) P-value

Treatment * 4.0 (1) 0.046unculled 15/133 (11.3 %) 13/74 (17.6 %) 0culled 0/91 (0 %) 0/58 (0 %) - 2.3

Year 0.4 (2) 0.82004 5/86 (5.8 %) 5/51 (9.8 %) 0 2005 6/60 (10 %) 6/52 (11.5 %) + 0.112006 4/78 (5.1 %) 3/47 (6.4 %) + 0.46

Age 6.5 (1) 0.01cub 5/173 (2.9 %) 5/105 (4.8 %) 0adult 10/51 (19.6 %) 8/39 (20.5 %) + 1.6

Sex 0.01 (1) 0.9male 8/115 (7.0 %) 7/66 (10.6 %) 0female 7/109 (6.4 %) 6/66 (9.1 %) - 0.05

* Model output for treatment based on running the model with one of the zero responses in the culled area replaced with 1 to make the algorithm tractable. All other outputs are from the raw data.

SID 5 (2/05) Page 21 of 31

Table 7 Results of a generalised linear mixed model (GLMM) predicting the probability of a positive ELISA result from live-sampling. Treatment (culled vs unculled), year, age and sex were fitted as main effects. Badger ID and sett were fitted as random effects.

Raw data Model outputvariable level n/N (%) ELISA

+ve trap eventsn/N (%) ELISA +ve individuals (per yr)

beta Wald (df) P-value

Treatment* 2.4 (1) 0.1unculled 58/133 (43.6 %) 40/76 (52.6 %) 0culled 23/96 (24.0 %) 15/59 (25.4 %) - 0.70

Year 0.1 (1) 0.972004 31/85 (36.5 %) 19/51 (37.3 %) 02005 25/61 (41.0 %) 20/52 (38.5 %) +0.062006 25/83 (30.1 %) 21/52 (40.4 %) + 0.08

Age 3.3 (1) 0.07cub 51/172 (29.7 %) 34/106 (32.1 %) 0adult 30/57 (52.6 %) 26/43 (55.8 %) + 0.63

Sex 0.5 (1) 0.5male 48/120 (40 %) 30/66 (45.5 %) 0female 33/109 (30.3 %) 24/67 (35.8 %) - 0.21

SID 5 (2/05) Page 22 of 31

* Model output for treatment based on running the model with one of the zero responses in the culled area replaced with 1 to make the algorithm tractable. All other outputs are from the raw data.

SID 5 (2/05) Page 23 of 31

Figure 13 Spatial distribution of culture positive badgers (solid grey circles), ELISA positive badgers (hatched circles) and badgers that tested negative for both (open circles) from live-sampling in the perturbation study area in 2004, 2005 and 2006. The size of the pies are proportional to the number of animals caught at that sett. Positive (solid squares) and negative (open squares) from post-mortem examination and tissue culture of RTAs are also shown. The dotted line shows the boundary of the proactive culling triplet I2.

SID 5 (2/05) Page 24 of 31

unculled culled

unculled culled

unculled culled

Figure 14 Spatial distribution of culture positive badgers (solid grey circles) and badgers that tested negative (open circles) from post-mortem examination and tissue culture during the 2004 and 2005 culling operations. The size of the pies are proportional to the number of animals caught at that location. Positive (solid squares) and negative (open squares) from post-mortem examination of RTAs are also shown. The dotted line shows the boundary of the proactive culling triplet I2.

Discussion

SID 5 (2/05) Page 25 of 31

unculled culled

unculled culled

This study provides evidence of several specific processes that together characterise social perturbation of badger populations after culling. The results are consistent with previous field studies showing that culling disrupts badger social organisation, demography and behaviour (Cheeseman et al. 1993; Roper & Lüps 1993; O’Corry-Crowe et al. 1996; Tuyttens et al. 2000a,b; Woodroffe et al. 2005a). The findings have significant implications for evaluating the sustainability of management of badger populations as a means of disease control.

Differences in disease prevalence between the treatment areas are consistent with the potential for culling to reduce TB prevalence in badger populations within the culled area, at least in the very short-term, and with the hypothesis that culling-induced perturbation of badger populations is associated with counter-productive effects in adjacent areas. These findings correspond with the pattern of infection found in cattle in culled and adjacent areas during the RBCT (Donnelly et al. 2003; Le Fevre et al. 2005; Donnelly et al. 2006).

The present study was not designed to address the full epidemiological consequences of culling and this study differs principally from the RBCT in its temporal scope. It was initiated in 2004, two years after the initial culling operation in I2 and one year after the first follow-up cull. While this study could not compare the effect of culling directly, the central aim was to compare the characteristics of a sample of badger social groups in an area subjected to sequential culling over several years with that of a sample in an adjacent unculled area. A major constraint on generalisation from this focused investigation could be lessened by wider investigations of other triplet areas of the RBCT.

Results from bait marking and radio-tracking were complementary, indicating that badgers in the area subjected to culling ranged further than those in the adjacent unculled area. Bait marking indicated a distinct difference in the spatial organisation of the two populations that varied over time. There was considerable variation in the size of social group ranges in the culled area and despite all of the active setts in the culled area being fed marked bait in 2005, uptake was low and bait returns were limited. Consequently, this measure of ranging behaviour almost certainly underestimated the true extent of badger social group ranges in the culled area.

Differences between treatment areas in the mean distance from setts to their associated bait returns are less susceptible to bias resulting from few returns. Mean bait return distances were greater for groups in the culled area than those in the unculled area in 2005 and 2007. Woodroffe et al. (2005a) detected a very similar difference between bait return distances inside and outside proactive culling areas of the RBCT one to two years after the final culling operation, suggesting that social groups in the culled population were extending their ranges. Woodroffe et al. (2005a) also reported similar differences between reactive culling areas and survey-only areas.

Badgers living in the area subjected to culling ranged more widely than those in the adjacent unculled area, as has been observed in other studies (Cheeseman et al. 1993; Roper & Lüps 1993; Defra report SE3108). This difference was most evident in the core ranges of badgers, and limited statistical power meant the differences in wider home ranges only approached significance. This distinction between core and wider home ranges may be important in respect to transmission rates. Core areas may represent areas of increased probability of disease transmission within groups, whereas home range areas might be expected to be more important for between group or between host transmission.

The magnitude of differences in the sizes of core areas and wider home ranges between culled and unculled areas increased between the third and fourth follow-up cull, and again after the fourth cull. If the increased ranging behaviour of badgers in the culled area was a direct consequence of culling operations this suggests that culling may influence the ranging behaviour of individuals for some time after disruption to social group organisation ceases to be detected through bait marking. This may be important because much of the evidence of social perturbation following culling has been collected by bait marking (Cheeseman et al. 1993; Tuyttens et al 2000a; Woodroffe et al. 2005a; Defra report SE3108), and may underestimate the degree of disruption at the level of the individual.

The relatively high proportion of cubs in the population observed throughout the study area is characteristic of other populations perturbed by culling (Tuyttens et al. 2000b; Defra report SE3108). The tendency for the cub:adult ratio to be higher in the culled area than in the adjacent unculled area in 2005 and 2006 is consistent with compensatory density-dependent reproduction (Lloyd-Smith et al. 2005). Badger populations have been found to have a higher potential productivity than that usually achieved, as although the majority of females carry blastocysts, only a minority subsequently implant and produce a litter (Neal & Cheeseman 1996). The proportion of females that implant has been found to decrease with increasing population density (Neal & Cheeseman 1996). Hence the reduction in population density caused by culling in the present study may have eased reproductive suppression amongst female badgers. This is a well-documented consequence of culling in other mammals (e.g. red foxes, Heydon & Reynolds 2000; brushtail possums, Ji et al. 2004) and some evidence of compensatory reproduction in badgers following culling has previously been reported (Tuyttens et al. 2000b). Compensatory reproduction may have the effect of increasing the long-term pool of susceptibles in the population with unpredictable consequences for disease persistence (Lloyd-Smith et al. 2005).

It is not clear why the cub:adult ratio was high in the unculled area in 2004 but decreased thereafter. There were a similar proportion of cubs to adults in the unculled area after 2004 as in the high-density, undisturbed population at Woodchester Park during the same period, suggesting that the ratio in the unculled area in 2005 and 2006 is more indicative of “normal” levels of productivity. Other studies have reported rapid immigration from neighbouring areas following culling (Cheeseman et al. 1993; Tuyttens et al. 2000a) and so the

SID 5 (2/05) Page 26 of 31

initially high proportion of cubs in the unculled area in 2004 may have been a result of the first two culling operations, if they resulted in a decreased density in the unculled area, followed by increased productivity.

An alternative hypothesis to explain the higher ratio of cubs to adults in the culled area is that culling may have left more neophobic or trap-shy adults. The anecdotal evidence gathered here by monitoring activity levels at setts prior to trapping, suggested that badgers living in the culled area responded markedly by avoiding areas where traps were set. If intervention in badger populations is associated with selection for neophobic traits this has implications not just for the success of culling operations, but also for other methods of TB control such as the delivery of vaccines in previously perturbed areas. A field experiment involving behavioural surveillance in culled and unculled areas could test this hypothesis.

Increases in the number of adult females relative to males in the culled area in 2005 and captures of more unmarked females in the culled area are consistent with the observation from a previous perturbation study that female badgers were the first to colonise culled areas (Tuyttens et al. 2000b). However, this does not explain the increase in the number of female cubs captured during the 2005 culling operation. This result contrasts with observations from the undisturbed population at Woodchester Park (SE3032), which has shown very little variation in the sex ratio between cubs over a 25-year period. A recent study of perturbation in a reactive culling area of the RBCT showed a similar trend for the adult sex ratio to become more female-biased following culling (Defra report SE3108), indicating that this may be a general pattern. Also, in the present study those badgers that were culled in 2005 exhibited an extremely skewed female biased sex ratio, whereas this was not the case amongst badgers culled in the previous study (SE3108).

Movements of badgers between social groups were very infrequent and usually involved male badgers (Rogers et al. 1998; Tuyttens et al. 2000b). A previous perturbation field study showed a short-term increase in extra-group movements in response to culling (Tuyttens et al. 2000b), but the small number of animals that moved between social groups in the present study prevented formal analysis.

Evidence from live sampling indicated substantial annual and spatial differences in the prevalence of animals excreting M. bovis. No positive cases were detected in the culled area. In the unculled area, however, the prevalence varied annually, reaching a peak in 2005. A similar, though less marked, pattern of lower prevalence in the culled area was observed from ELISA results. Intriguingly, the prevalence of ELISA positive cases dropped substantially from 39% to 5% in the culled area in 2005 and increased to 26% in the following year. This pattern is reflected in the prevalence of culture positive cases determined from post mortem examination of culled animals for the two years that this study and the culling operations overlapped.

Critically, the peak in positive cases observed in the unculled area in 2005 is coincident with the most marked differences in ranging behaviour indicated by both bait marking and radio-tracking. In particular, the two largest social group ranges in the culled area were neighbouring or overlapping the boundary between the culled and unculled areas, providing potential for movement of badgers and increased risk of transmission between the two areas.

Prevalence estimates calculated from live-sampling and from post mortem examination of culled badgers differed from each other. There are three possible explanations for this observation, either: 1. the live-test was too insensitive to detect M. bovis in badgers found later to be culture positive at post mortem; 2. they were intermittently excreting bacilli and all were latent at the time of the live-sampling; 3. disease progression (culture negative to culture positive) had occurred between the live-sampling event and the animal being killed. Of these, the first possibility that all were false negatives and the second that latency among several animals coincided both seem improbable. Given the time interval between live-sampling and animals being killed, and the observation that animals testing positive at post mortem were not in an advanced stage of disease, the third explanation seems most likely. Nethertheless, it is probable that this was further confounded by insensitivity of live-sampling (Pritchard et al. 1986) and intermittent excretion (Gallagher & Clifton-Hadley 2000).

In conclusion, this study provides further evidence for culling-induced perturbation of badger social organisation. Uniquely however, these results relate to a population that was subjected to sequential annual culling and suggest that the processes of disturbance persisted at least as long as culling continued and varied in their extent over this period. Although culling appeared to have a beneficial effect on levels of detectable infection in badgers in the culled area, infection levels in the adjacent area peaked when levels of disturbance were greatest. These findings have direct relevance for assessing the sustainability of a range of policies for controlling bovine TB.

7. Future work and actions arising A SID3 has already been submitted detailing a proposal to continue investigations into culling-induced perturbation of the population in I2, consistent with that described in this report. In addition, it contains a proposal to investigate experimentally and attempt to quantify anecdotal observations of increased levels of trap and bait shyness in a badger population subjected to culling. This proposal is currently under review. A constraint on the generalisation of results from this intensive study is the lack of replication. A more extensive study to assess the sustainability of culling by measuring some of the most important parameters related to social perturbation in I2 and other triplet areas of the RBCT would enable greater generalisation of results.

SID 5 (2/05) Page 27 of 31

The perturbation study area provides a complementary low-density, disturbed population with which to compare investigations of a range of potential management options to control M. bovis. Defra has identified the development of TB vaccines for badgers as a potentially important area for current research. It is likely that a vaccine would need to be delivered in an oral bait. Bait uptake studies will be taking place during 2007 and 2008 in the high-density population at Woodchester Park. The intensively studied badger population in the perturbation study area provides a contrasting population to Woodchester to develop and test techniques to assess the proportion of a badger population that takes an oral bait. The population in I2 is representative of many previously disturbed areas of high incidence of TB in cattle, where a vaccine would most likely be targeted.

This established study area could also be used to fully evaluate novel methods of estimating population size and contact rates between badgers, neither of which have been assessed in disturbed, low-density populations. Proximity tags fitted to radio-collars have already been tested at Woodchester Park and the optimisation of this technique for assessing transmission risk between badgers would benefit from a low density, culled population.

An in depth analysis of the demographic and associated epidemiological data collected throughout the RBCT, in comparison with data collected during the perturbation study in I2, would be useful to identify individual characteristics of culls in relation to their success. Comparison of the data collected for this study with RBCT data showed that the majority of cubs culled in the 2004 culling operation were incorrectly classified as adults due to the time of year the cull was carried out. Assuming this occurred in other culling operations in the RBCT it would be possible to correctly re-classify age classes of RBCT culled badgers in the majority of cases using the Woodchester Park dataset. Thus enabling a meaningful analysis as suggested above to be carried out.

The results presented in this report are currently being compiled for submission to peer reviewed journals. A more detailed analysis of data collected in this study will be conducted in conjunction with complementary data from Woodchester Park and from the RBCT as they become available.

8. References

Bourne, F.J., Donnelly, C., Cox, D., Gettinby, G., McInerney, J., Morrison, I. & Woodroffe, R. 1998 Towards a sustainable policy to control TB in cattle. A Scientific Initiative. First Report of the Independent Scientific Group on Cattle TB. London: Ministry of Agriculture, Fisheries & Food.

Cheeseman, C.L. & Mallinson, P.J. 1979 Radio tracking in the study of bovine tuberculosis in badgers. In Amlaner Jr, C.J. & Macdonald, D.W. (eds). A Handbook on Biotelemetry and radio tracking 649-656. Pergamon Press, Oxford and New York.

Cheeseman, C.L. & Harris, S. 1982 Methods of marking badgers (Meles meles). Journal of Zoology 197, 289-292.

Cheeseman, C.L., Cresswell, W.J., Harris, S. & Mallinson, P.J. 1988 Comparison of dispersal and other movements in two badger Meles meles populations. Mamm. Rev. 18, 51-59.

Cheeseman, C.L., Mallinson, P.J., Ryan, J. & Wilesmith, J.W. 1993 Recolonisation by badgers in Gloucestershire. In The Badger (ed T.J. Hayden), pp. 78-93. Dublin: Royal Irish Academy.

Cochran, W.W. 1980 Wildlife telemetry. In S.D. Schemnitz (ed). Wildlife Management Techniques Manual, 4th

edition. 507-520. Wildlife Society, Washington.

Delahay, R.J., Brown, J.A., Mallinson, P.J., Spyvee, P.D., Handoll, D., Rogers, L.M. & Cheeseman, C.L. 2000 The use of marked bait in studies of the territorial organisation of the European badger Meles meles. Mamm. Rev. 30, 73-87.

Delahay, R.J., Brown, J.A., Mallinson, P.J., Spyvee, P.D., Handoll, D., Rogers, L.M. & Cheeseman, C.L. 2000a The use of marked bait in studies of the territorial organization of the European badger (Meles meles). Mammal Review 30(2), 73-87

de Leeuw, A. N., Forrester, G., Spyvee, P.D., Brash, M. & Delahay, R.J. 2004. Experimental comparison of ketamine with a combination of ketamine, butorphanol and medotomidine for general anaesthesia of the Eurasian badger (Meles meles L.). Veterinary Journal 167, 186-193.

Donnelly, C.A., Woodroffe, R., Cox, D.R., Bourne, J., Gettinby, G., Le Fevre, A.M., Mclnerney, J.P. & Morrison, W.I. 2003 Impact of localized badger culling on tuberculosis incidence in British cattle. Nature 426, 834-837.

SID 5 (2/05) Page 28 of 31

Donnelly, C.A., Woodroffe, R., Cox, D.R., Bourne, J., Cheeseman, C.L., Clifton-Hadley, R.S., Gettinby, G., Gilks, P., Johnston, W.T., Le Fevre, A.M., McInerney, J.P. & Morrison, W.I. 2006 Positive and negative effects of widespread badger culling on tuberculosis in cattle. Nature 439, 843-846.

Environmental Systems Research Institute (ESRI) 2005 ArcGIS, version 9.1 California: ESRI.

Eves, J.A. 1999 Impact of badger removal on bovine tuberculosis in east County Offaly. Ir. Vet. J. 52, 199-203.

Gallagher, J. & Clifton-Hadley, R.S. 2000 Tuberculosis in badgers; a review of the disease and its significance for other animals. Res. Vet. Sci. 69, 203-217.

Goodger, J., Nolan, A., Russell, W.P., Dalley, D.J., Thorns, C.J., Stuart, F.A., Croston, P. & Newell, D.G.1994. Serodiagnosis of Mycobacterium bovis infection in badgers: development of an indirect ELISA using a 25 kDa antigen. Veterinary Record 135, 82-85.

Griffin, J.M., Williams, D.H., Kelly, G.E., Clegg, T.A., O’Boyle, I.O., Collins, J.D. & More, S.J. 2005 The impact of badger removal on the control of tuberculosis in cattle herds in Ireland. Prev. Vet. Med. 67, 237-266.

Heydon, M.J. & Reynolds, J.C. 2000 Demography of rural foxes (Vulpes vulpes) in relation to cull intensity in three contrasting regions of Britain. J. Zool. 251, 265-76.

Ji, W., Sarre, S.D., White, P.C.L., Clout, M.N. 2004. Population recovery of common brushtail possums after local depopulation. Wildlife Research, 31, 543-550.

Kenward, R.E., South, A.B. & Walls, S.S. 2003 Ranges 6, Version 1.2: For the analysis of tracking and location data. Anatrack Ltd., Wareham, UK.

Le Fevre, A.M., Donnelly, C.A., Cox, D.R., Bourne, F.J., Clifton-Hadley, R.S., Gettinby, G., Johnston, W.T., McInerney, J.P., Morrison, W.I. & Woodroffe, R. 2005 The Impact of Localised Reactive Badger Culling Versus No Culling on TB Incidence in British cattle: a Randomised Trial. (http://www.defra.gov.uk/animal/tb/isg/pdf/ lefevre1005.pdf.)

Lloyd-Smith, J.O., Cross, P.C., Briggs, C.J., Daugherty, M., Getz, W.M., Latto, J., Sanchez, M.S., Smith, A.B. & Swei, A. 2005 Should we expect population thresholds for wildlife disease? Trends Ecol. Evolut. 20, 511-519.

Macdonald, D.W., Riordan, P. & Mathews, F. 2006 Biological hurdles to the control of TB in cattle: a test of two hypotheses concerning wildlife to explain the failure of control. Biol. Cons. 131, 268-286.

Neal, E. & Chesseman, C.L. 1996 Badgers, London: Poyser

O’Corry-Crowe, G., Eves, J. & Hayden, T.J. 1993 Sett distribution, territory size and population density of badgers (Meles meles L.) in East Offaly. In The Badger (ed T.J. Hayden), pp. 35-56. Dublin: Royal Irish Academy.

O’Corry-Crowe, G., Hammond, R., Eves, J. & Hayden, T.J. 1996 The effect of reduction in badger density on the spatial organisation and activity of badgers Meles meles L. in relation to farms in central Ireland. Biology and Environment: Proceedings of the Royal Irish Academy 96B, 147-158.

Pritchard, D.G., Stuart, F. A., Wilesmith, J.W., Cheeseman, C.L., Brewer, J.I., Bode, R. & Sayers, P.E. 1986 Tuberculosis in East Sussex. III. Comparison of post-mortem and clinical methods for the diagnosis of tuberculosis in badgers. J. Hyg. 97, 27-36

Rogers, L.M., Cheeseman, C.L., Mallinson, P.J. & Clifton-Hadley, R. 1997. The demography of a high density badger (Meles meles) population in the west of England. Journal of Zoology 242, 705-728..

Rogers, L.M., Delahay, R., Cheeseman, C.L., Langton, S., Smith, G.C. & Clifton-Hadley, R.S. 1998 Movement of badgers Meles meles in a high-density population: individual, population and disease effects. Proc. R. Soc. B 265, 1269-1276.

Roper, T.J. & Lüps, P. 1993 Disruption of territorial behaviour in badgers Meles meles. Z. Säuget. 58, 252-255.

Seaman, D.E., Millspaugh, J.J., Kernohan, B.J., Raedeke, K.J. & Gitzn, R.A., 1999. Effects of sample size on kernel home range estimates. Journal of Wildlife Management, 63, 739-747.

Tuyttens, F.A.M., Delahay, R.J., Macdonald, D.W., Cheeseman, C.L., Long, B. & Donnelly, C.A. 2000a Spatial perturbation caused by a badger Meles meles culling operation: implications for the function of territoriality and

SID 5 (2/05) Page 29 of 31

the control of bovine tuberculosis Mycobacterium bovis. J. Anim. Ecol. 69, 815-828.

Tuyttens, F.A.M., Macdonald, D.W., Rogers, L.M., Cheeseman, C.L. & Roddam, A.W. 2000b Comparative study on the consequences of culling badgers (Meles meles) on biometrics, population dynamics and movement. J. Anim. Ecol. 69, 567-580.

Vicente, J., Delahay, R.J., Walker, N. & Cheeseman, C. L. 2007 Social organization and movement influence the incidence of bovine tuberculosis in an undisturbed high density badger Meles meles population. J. Anim. Ecol. (DOI 10.111/j.1365-2656.2006.01199.x)

Woodroffe, R., Donnelly, C.A., Cox, D.R., Bourne, J.F., Cheeseman, C.L., Delahay, R.J., Gettinby, G., McInerney, J.P. & Morrison, W.I. 2005a Effects of culling on badger Meles meles spatial organization: implications for the control of bovine tuberculosis. J. Appl. Ecol. 42, 852-862.

Woodroffe, R., Bourne, F.J., Cheeseman, C.L., Cox, D.R., Donnelly, C.A., Gettinby, G., McInerney, J.P. & Morrison, W.I. 2005b. Welfare of badgers (Meles meles) subjected to culling: development and evaluation of a closed season. Animal Welfare 14, 19-25.

Woodroffe, R., Donnelly, C.A., Jenkins, H.E., Johnston, W.T., Cox, D.R., Bourne, F.J., Cheeseman, C.L., Delahay, R.J., Clifton-Hadley, R.S., Gettinby, G., Gilks, P., Hewinson, R.G., McInerney, J.P. & Morrison, W.I. 2006 Culling and cattle controls influence tuberculosis risk for badgers. Proc. Natl. Acad. Sci. USA. 103, 14713-14717. (DOI 10.1073/pnas.0606251103.)

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

SID 5 (2/05) Page 30 of 31

SID 5 (2/05) Page 31 of 31