SEARCH

SEARCH BY CITATION

Keywords:

  • bait marking;
  • bovine TB;
  • European badger;
  • Krebs trial;
  • Mycobacterium bovis;
  • perturbation;
  • randomized badger culling trial;
  • reservoir host;
  • wildlife disease;
  • zoonosis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    The incidence of bovine tuberculosis (TB) in British cattle has risen markedly over the last two decades. Failure to control the disease in cattle has been linked to the persistence of a reservoir of infection in European badgers Meles meles, a nationally protected species. Although badger culling has formed a component of British TB control policy for many years, a recent large-scale randomized field experiment found that TB incidence in cattle was no lower in areas subject to localized badger culling than in nearby areas where no experimental culls occurred. Indeed, analyses indicated that cattle incidence was higher in culled areas.
  • 2
    One hypothesis advanced to explain this pattern is that localized culling disrupted badgers’ territorial behaviour, potentially increasing the rate of contact between cattle and infected badgers. This study evaluated this hypothesis by investigating badger activity and spatial organization in 13 study areas subjected to different levels of culling. Badger home ranges were mapped by feeding colour-marked baits at badger dens and measuring the geographical area in which colour-marked faeces were retrieved.
  • 3
    Badger home ranges were consistently larger in culling areas. Moreover, in areas not subjected to culling, home range sizes increased with proximity to the culling area boundary. Patterns of overlap between home ranges were also influenced by culling.
  • 4
    Synthesis and applications. This study demonstrates that culling badgers profoundly alters their spatial organization as well as their population density. These changes have the potential to influence contact rates between cattle and badgers, both where culls occur and on adjoining land. These results may help to explain why localized badger culling appears to have failed to control cattle TB, and should be taken into account in determining what role, if any, badger culling should play in future control strategies.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The management of European badgers Meles meles L. is of great concern in the British Isles, where this species is implicated in the transmission of bovine tuberculosis (TB) to cattle (Krebs et al. 1997; Eves 1999). Bovine TB, caused by the bacterium Mycobacterium bovis, is a zoonotic disease of cattle that has been increasing in incidence since the 1980s (Krebs et al. 1997).

Over the last 30 years, policies to control cattle TB in Britain have included two main components. The first component is intended to detect infection in cattle and prevent spread to other herds, and involves regular testing of cattle, with slaughter of any animals showing evidence of possible infection, as well as temporary movement restrictions on affected herds. Cattle control measures similar to these have successfully eradicated TB in much of the developed world. Control has been more difficult to achieve, however, where wildlife populations are persistently infected (Morris, Pfeiffer & Jackson 1994). Hence, the second component of British TB control policy has involved various forms of badger culling, intended to reduce the risk of transmission to cattle in high-risk areas (Zuckerman 1980; Dunnet, Jones & McInerney 1986; Krebs et al. 1997). Badgers and their dens (setts) are protected by national legislation and, since the 1970s, all badger culling carried out for disease control purposes has been performed by government staff.

Although various forms of badger culling have formed a component of British TB policy since 1973 (Krebs et al. 1997), their effectiveness as control measures appears variable. Virtual elimination of badgers from several localities in the British Isles has been linked to declines in the incidence of TB in cattle (Clifton-Hadley et al. 1995; Eves 1999; Griffin et al. 2005). Spatial associations between patterns of M. bovis infection in cattle and badgers indicate that transmission occurs between the two species at a local scale (Woodroffe et al. 2005), suggesting that more localized culling of badgers could also be expected to control cattle TB. However, analysis of data from a large-scale randomized, replicated and controlled field experiment (the Randomized Badger Culling Trial, RBCT) recently showed that the incidence of TB in cattle was higher in areas subjected to localized badger culling, similar to the form of culling implemented as past policy in much of Britain and Ireland, than in nearby areas where no experimental culling occurred (Donnelly et al. 2003; Le Fevre et al. 2005). The British government suspended experimental localized badger culling in 2003, because its failure to reduce the incidence of cattle TB on the time-scale tested indicated that this approach could be expected to contribute little to improved control (Donnelly et al. 2003; Bourne et al. 2005). Subsequently, however, similar management approaches have been included in policy proposals put forward by lobby groups (British Veterinary Association 2005; Gallagher et al. 2005; National Farmers’ Union 2005). A better understanding of the mechanisms underlying the link between badger culling and TB in cattle is vital to determine whether culling can be expected to contribute more effectively to TB control, or whether other management actions (e.g. badger vaccination, fertility control or improved cattle controls; White & Harris 1995; Swinton et al. 1997) might be more effective.

One explanation for the possibly detrimental effect of localized badger culling is that the consequent disruption of badger spatial organization might influence TB transmission, either among badgers or from badgers to cattle (Swinton et al. 1997; Tuyttens et al. 2000; Tuyttens & Macdonald 2000; Donnelly et al. 2003). Frequent movement of badgers between social groups has been associated with increased transmission of TB within a badger population (Rogers et al. 1998), and such movements are known to occur more frequently in low density badger populations, such as those suppressed by culling (Woodroffe, Macdonald & da Silva 1995). Hence, reduction of population density may reduce contact rates between badgers to a lesser extent than expected (Barlow 1996), and could even increase them (Swinton et al. 1997), potentially increasing TB prevalence in the badger population. In addition, badger culling has been linked to increases in the extent of badger ranging behaviour (Cheeseman et al. 1993; O’Corry-Crowe et al. 1996; Tuyttens et al. 2000), raising the possibility that individual badgers in populations subject to control may come into contact with a larger number of cattle herds and, if infected, could potentially trigger a larger number of infections in cattle. In TB-infected badger populations, either or both of these processes could increase the risk of transmission from badgers to cattle. Importantly, such ‘perturbation’ mechanisms may interact with simpler effects of reduced badger population density to produce TB dynamics that are strongly non-linear. Hence, they could reconcile the apparently contradictory findings that elimination of badgers appears to contribute to the control of cattle TB (Clifton-Hadley et al. 1995; Eves 1999; Griffin et al. 2005), while localized culling may have little effect or even increase TB risks to cattle (Donnelly et al. 2003).

An alternative, somewhat simpler, explanation for the apparently detrimental effects of localized culling is that experimental control areas could have been compromised (Godfray et al. 2004). Specifically, if landowners in areas randomly allocated to receive no officially sanctioned culling were, in fact, culling badgers illegally, then badger density might be lower in these areas than in those that received localized culling. If this was the case, it could explain why legal, government-implemented, culling appears to be less effective at controlling cattle TB than no (legal) culling at all.

Only limited information is available to evaluate the hypothesis that culling prompts perturbation of badger spatial organization in ways that might influence TB dynamics. Past studies have included before and after comparisons at single sites in response to localized culling (Cheeseman et al. 1993; O’Corry-Crowe et al. 1996; Tuyttens et al. 2000) as well as extrapolation of patterns from an undisturbed population (Rogers et al. 1998). Only one study has simultaneously compared two sites with and without localized culling (P. Riordan, D.W. Macdonald, R.J. Delahay, C.L. Cheeseman, K.M. Service, E. Fordham & B.J. Harmsen, unpublished data). Hence, the only studies of the effects of localized culling, similar to that implemented in the RBCT, document one-off changes in badger spatial organization, which might be because of site-specific factors. Given the need to understand the importance, if any (Bourne et al. 2004; Godfray et al. 2004), of social perturbation in influencing TB dynamics and, hence, the expected efficacy of measures such as badger vaccination in comparison with culling, there is a clear need for replicated studies that investigate general patterns of change in badger spatial organization in response to culling. To evaluate the hypothesis that culling may disrupt badger spatial organization, we compared the activity and spatial organization of badgers in 13 study areas exposed to different levels of culling.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

study areas

We carried out this study in 13 of the 30 trial areas that were enrolled in the RBCT. Details of the RBCT design and justification are given in Bourne et al. (1998, 1999) and Woodroffe et al. (2002). Briefly, the 30 100-km2 trial areas were grouped into 10 ‘triplets’. In each triplet, one trial area was randomly assigned to the proactive culling treatment (which aimed to reduce badger densities to very low levels across the entire trial area), one to the reactive culling treatment (localized culling, which sought to remove only those badgers whose home ranges overlapped farms experiencing cattle TB outbreaks) and a third to the no-culling ‘survey-only’ treatment. All government-sanctioned culling was restricted to trial areas, with no legal culling outside.

This study took place in Triplet B on the Devon–Cornwall border (approximately 4°26′W, 50°54′N), Triplet C in East Cornwall (approximately 4°32′W, 50°22′N), Triplet D in Herefordshire (approximately 2°27′W, 52°4′N), Triplet G on the Staffordshire–Derbyshire border (approximately 1°52′W, 53°4′N) and Triplet H on the Devon–Somerset border (approximately 3°29′W, 51°7′N). These triplets were chosen to represent a range of badger culling histories prior to the start of the RBCT (Table 1), from Triplet G, which had no prior culling by government staff, to Triplet B, which had been culled repeatedly.

Table 1.  Culling history of the study areas in five triplets. ‘Last culled’ denotes the year in which badgers were most recently removed from the study area prior to the bait-marking operations; the reactive treatment was discontinued in 2003 but additional proactive culls occurred in Triplets G and H following the 2004 surveys. ‘Years culled’ indicates the number of calendar years in which one or more badgers were removed from the study area; these years were not necessarily consecutive
TripletTreatmentYear surveyedRBCT (1998–2004)Badger removal operations (1986–98)
Last culledYears culledBadgers culledLast culledYears culledBadgers culled
Binside proactive2005200461141998880
outside proactive20050  01997527
reactive200520033 381996415
survey-only20050  01997422
Cinside proactive2005200451391998270
outside proactive20050  01997126
Dinside proactive2005200432410 0
outside proactive20050  00 0
reactive200520031 160 0
survey-only20050  01997111
Ginside proactive2004200332820 0
outside proactive20040  00 0
reactive200420021 260 0
survey-only20040  00 0
Hinside proactive2004200331720 0
outside proactive20040  00 0
reactive200420031 831998359
survey-only20040  019961 6

We nominated a study area in each of 12 trial areas comprising four triplets. Study areas in the reactive (B1, D1, G1, H1) and survey-only (B3, D2, G3, H3) areas each covered approximately 16 km2 (4 × 4 km) falling completely within their respective trial areas. Study areas in the proactive areas of these four triplets (B2, D3, G2, H2), as well as a thirteenth study area in the proactive area of Triplet C (C3), each covered approximately 24 km2 (6 × 4 km), and were positioned so that approximately 16 km2 (4 × 4 km) fell inside the culling areas and about 8 km2 (2 × 4 km) fell outside, on land that was not culled in the course of the RBCT (Fig. 1). Proactive study areas were placed in this way to explore possible effects of widespread culling on badger populations in adjoining areas. Because of limited resources, no study areas were designated in the reactive or survey-only trial areas of Triplet C.

image

Figure 1. Spatial organization of badgers in the survey-only (a), reactive (b) and proactive (c) study areas of Triplet D. Stars show the location of bait-marked setts, open circles indicate latrines and solid lines denote minimum convex polygons around bait returns. Dotted lines show the boundaries of study areas; the heavy dashed line in the proactive study area shows the boundary of the culling area. Shading indicates the division of the study areas into sectors for the purposes of statistical analysis. Similar maps of the other 10 study areas are available in Figs S1–S4 in the Supplementary Material.

Download figure to PowerPoint

bait marking

Badger spatial organization was investigated by bait marking, carried out in accordance with Kruuk (1978) and Delahay et al. (2000). This technique was used in preference to radio-telemetry because it allowed simultaneous mapping of badger home ranges across multiple areas at greatly reduced cost. In brief, each area was thoroughly surveyed for signs of badger activity, particularly setts, and latrines at which badgers regularly scent marked using faeces, urine and various glandular secretions (Kruuk 1978). Probable ‘main setts’, dens that are occupied year round and are the centres of activity for territorial social groups (Neal & Cheeseman 1996), were identified on the basis of their size and activity (e.g. evidence of fresh digging, bedding, tracks and faeces). Bait (peanuts mixed with treacle) containing indigestible colour markers (small plastic beads) was fed to badgers at all suspected main setts. Multiple baits were laid at badger setts, and replenished daily, for a 12-day period, using a different colour marker for each sett. Immediately following the feeding period, each study area was resurveyed, seeking evidence of faecal deposits containing the colour markers; these were termed ‘bait returns’. The location of each bait return was mapped, and the number of latrine pits containing marked faeces was recorded to give a measure of the number of bait returns per sett. Bait marking was carried out approximately simultaneously in Triplets G and H during 2004 (G, March–April 2004; H, February–March 2004) and in Triplets B, C and D during 2005 (B and C, February–March 2005; D, March–April 2005). The locations of badger setts, latrines and all bait returns were mapped using geographical information system (GIS) software (ArcView, Environmental Systems Research Institute Inc., Redlands, CA).

data analysis

We compared the number and spatial distribution of bait returns across culling treatments. As each proactive study area spanned the boundary of a culling area, we distinguished setts that fell inside the culling areas from those that fell outside. Hence, comparisons across treatments contained four types of sett: those inside the proactive culling area, those immediately outside this area, those in the reactive culling area and those in the survey-only area.

The population density and ranging behaviour of badgers is known to vary with ecological conditions, at both national and local scales (Kruuk & Parish 1982; Woodroffe & Macdonald 1993). This indicates that data gathered from badger social groups within a triplet would be expected to be more similar to one another, all else being equal, than to data from other triplets. Within trial areas, data from social groups located in close proximity to one another are likewise expected to be similar because such groups are likely to occupy home ranges with similar habitat characteristics, and may also be genetically related. These expected similarities mean that data from setts, social groups and home ranges located in close proximity to one another are unlikely to be independent. We accounted for similarities within triplets by including triplet as a covariate in all analyses performed. To account for more localized intercorrelation, we divided each study area into sectors (Fig. 1), and summarized data from each sector for the purposes of analysis. Proactive study areas were each divided into four sectors, two inside and two outside the culling area boundary, with dividing lines between sectors approximately parallel to this boundary. Study areas in survey-only and reactive culling areas were divided into two sectors of approximately equal size, with dividing lines between sectors likewise roughly parallel to the nearest treatment area boundary. Unless otherwise stated, the units of analysis for all subsequent investigations were the medians for these sectors (e.g. the median number of bait returns per sett). One sector outside the Triplet G proactive culling area contained no setts; hence the total number of sectors analysed was 35. The median number of setts per sector was eight (range 3–18). Comparisons within and between study areas were performed using generalized linear models, including a term for triplet in each model. All mean values presented are least squares means, which estimate what the means would have been had complete data been available (i.e. with data from the last sector of Triplet G and from the survey-only and reactive culling areas of Triplet C). As a check, we also repeated the primary analyses excluding data from Triplet C; this produced a balanced study design but also reduced the sample size to 31 sectors.

We delineated badger social group ranges by drawing minimum convex polygons around the outermost bait returns for each sett (Fig. 1). In high-density populations, such polygons have been shown to provide a good approximation to the territories of badgers estimated from radio-telemetry (Kruuk 1978; Delahay et al. 2000). Following Delahay et al. (2000), we excluded from delineated home ranges a small number of bait returns (0·6% of 6445 returns) that were located deep within the home ranges of other social groups (in most cases, single returns located beyond the other group's main sett but with no further evidence that a home range was shared between badgers from more than one sett). Such returns, which are detected on most bait-marking exercises (Delahay et al. 2000), probably reflect infrequent ‘excursions’ to neighbouring setts by badgers of both sexes in search of mating opportunities (Woodroffe, Macdonald & da Silva 1995); the bait-marking exercises were carried out during the mating season, which is when the peak of territorial activity occurs (Roper, Sheperdson & Davies 1986). The areas of these minimum convex polygons gave one measure of badgers’ ranging behaviour. However, after accounting for treatment effects (non-significant triplet effects excluded from the model), this measure of home range size (expressed as a median value for each sector) was positively correlated with the median number of bait returns per sett (F1,30 = 38·46, P < 0·001; n= 35 sectors). Hence this measure potentially underestimates ranging behaviour where the number of bait returns was low. We therefore calculated an alternative measure of badgers’ ranging behaviour: the median distance between a sett and its associated bait returns. This measure, expressed as a median value for each sector, was not significantly correlated with the number of bait returns per sett (F1,30 = 0·18, P= 0·672; n= 35 sectors). Hence, we used this measure to compare badgers’ ranging behaviour across treatments.

We were concerned that a proportion of bait returns would be missed by survey teams if they came from setts close to the boundary of the area surveyed. To investigate the possible impact of such ‘edge effects’ on our primary analyses, we divided each survey-only area and each reactive study area into ‘core’ and ‘peripheral’ zones with equal area. We restricted these analyses to survey-only and reactive areas because, in proactive areas, culling occurred at one end of each study area, and not at the other. This could make it difficult to distinguish edge effects from effects of culling; we therefore excluded proactive areas from this analysis. In the survey-only and reactive areas, after adjusting for triplet and treatment effects the difference in the median number of bait returns per sett recorded in core and peripheral areas was not consistent (in magnitude or direction) across triplets, giving rise to a significant triplet–zone interaction (F3,7 = 12·76, P= 0·003; n= 26 zones). Neither the median distance between setts and bait returns, nor the median area of minimum convex polygons, varied between zones (median distances, F1,10= 0·35, P= 0·569; areas, F1,10 = 1·51, P= 0·248; n= 16 zones) and there were no significant interactions. Hence we were reasonably confident that our measures of ranging behaviour were not biased by edge effects.

Bait marking indicated that a proportion of home ranges contained more than one large sett, identified on the initial surveys as possible main setts (Fig. 1). This is commonly encountered in bait-marking studies (Delahay et al. 2000). We used the proportion of home ranges that contained two or more large setts, and the distance between these setts, as additional indices of badger spatial organization. We also measured the proportion of each home range polygon that overlapped with other home ranges, and the number of other home ranges overlapping each range. On the few occasions when such shared home ranges spanned sector boundaries, with a sett either side, home ranges were allocated to the sector containing the sett with the largest number of bait returns, as these were judged to be the most probable main setts.

Primary analyses compared bait returns and home range characteristics across the four culling treatments. As a second analysis, we also compared the proportion of setts within the reactive culling areas that either had, or had not, experienced culling in the course of the RBCT. As not all badgers were captured at main setts (that were bait marked), and as the level of activity of particular setts would in any case be expected to vary in response to culling (possibly altering the designation of likely main setts), we considered bait-marked setts to have experienced culling if any badgers were taken within a 500-m radius. This radius was chosen to approximate the expected home range size of badgers occupying British farmland (Woodroffe & Macdonald 1993). Within proactive study areas, we also analysed relationships between bait-marking results and distances from setts to the boundaries of the culling areas.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

bait returns per sett

The number of bait returns per sett, expressed as the median value for each sector, varied with the treatment applied (F3,27 = 5·51, P= 0·004; n= 35 sectors), with no significant effect of triplet (F4,27 = 1·83, P= 0·152; the non-significant triplet*treatment interaction was excluded from the final model). As expected, bait returns were highest for areas that were not culled (survey-only areas and the unculled areas outside proactive areas) and lowest for culled areas (Fig. 2a). Very similar results were obtained when data from the triplet C proactive area were excluded (treatment effect F3,24= 5·20, P= 0·007; triplet effect F3,24= 2·21, P= 0·113; the non-significant triplet*treatment effect excluded; n= 31 sectors).

image

Figure 2. Differences between culling treatments in (a) the number of marked faecal deposits (bait returns) retrieved per sett where colour-marked baits were placed; (b) the median distance (per sett) from setts to bait returns; and (c) the number of other home ranges (HRs) overlapping each mapped range. All figures are least-squares means, averaged across median values for each sector, and error bars denote standard errors.

Download figure to PowerPoint

In reactive culling areas, the median number of returns per sett was unrelated to the proportion of setts within a sector that had received badger culling (F1,3 = 0·28, P= 0·635; n= 8 sectors).

Within the proactive study areas, the number of bait returns per sett was related to the distance from the treatment area boundary (Fig. 3a). After adjusting for triplet, the effect of the distance from the boundary of the culling area was significantly different (F1,12 = 8·56, P= 0·013; n= 19 sectors) inside and outside the culling areas. Outside the culling areas, the number of returns was significantly lower for setts located closer to the area culled (slope =−0·018 m−1, SE = 0·004), but inside the culling area the slope (−0·002 m−1, SE = 0·002) was not significantly different from zero (Fig. 3a).

image

Figure 3. Relationships between bait-marking results and distance from the boundary of the proactive culling area (indicated by a dashed line), showing (a) bait returns per sett and (b) the median distance from setts to their associated bait returns. All values represent medians for each sector. Lines indicate statistically significant slopes outside proactive culling areas; slopes inside culling areas were not significantly different from zero and are not shown.

Download figure to PowerPoint

bait return distance

The median distance from setts to bait returns, expressed as the median value for each sector, varied with the treatment applied (F3,27 = 6·99, P= 0·001; n= 35 sectors), with no significant effect of triplet (F4,27 = 0·74, P= 0·575; the non-significant triplet*treatment interaction was excluded from the final model). Distances were smallest for areas that were not culled (survey-only areas and the unculled areas outside proactive areas) and highest for culled areas (Fig. 2b). Very similar results were obtained when data from the triplet C proactive area were excluded (treatment effect F3,24= 5·09, P= 0·007; triplet effect F3,24= 0·88, P= 0·465; the non-significant triplet*treatment effect excluded; n= 31 sectors).

Within reactive culling areas, the median distance between setts and their associated bait returns was unrelated to the proportion of setts within a sector that had received badger culling (F1,3 = 0·13, P= 0·742; n= 8 sectors).

Within the proactive study areas, the median bait return distance was related to the distance from the treatment area boundary (Fig. 3b). After adjusting for triplet, the effect of distance from the culling area boundaries (expressed as the median for each sector) was significantly different inside and outside the culling areas (F1,12 = 8·58, P= 0·013; n= 19 sectors). Outside the culling areas, the median return distance was significantly higher in sectors located closer to the area culled (slope = 0·244 m−1, SE = 0·071), but inside the culling area the slope (−0·030 m−1, SE = 0·032) was not significantly different from zero (Fig. 3b).

home range overlap

Patterns of overlap between mapped home ranges also varied between treatments. After adjusting for triplet, the median number of other home ranges overlapping each range varied significantly between culling treatments (F3,27 = 3·54, P= 0·028; the non-significant triplet* treatment interaction was excluded from the final model; n= 35 sectors). Very similar results were obtained when data from the triplet C proactive area were excluded (treatment effect F3,24= 3·51, P= 0·031; non-significant triplet*treatment effect excluded; n= 31 sectors).

Overlap was greatest both outside and inside proactive culling areas, with fewer overlaps in reactive and survey-only areas (Fig. 2c). There was no significant effect of culling treatment on the median proportion of each home range that overlapped with other ranges (F3,27 = 1·21, P= 0·324).

home ranges with multiple main setts

In 12 of the 13 study areas, a proportion of mapped home ranges contained more than one large sett identified as a likely main sett and consequently bait marked. Comparisons across the four treatments showed that the proportion of home ranges containing multiple bait-marked setts was significantly influenced by the triplet*treatment interaction (F10,17= 2·70, P= 0·035; n= 35 sectors). The median distance between such linked setts was likewise affected by this interaction term (F6,7 = 7·43, P= 0·009; n= 21 sectors; note that the sample size for this analysis was smaller because only 21 of the 35 sectors contained linked setts). These significant interactions indicate different effects of the culling treatments in different triplets, making it impossible to draw conclusions about general trends.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Our data indicate that culling influenced both the activity and spatial organization of badgers. The numbers of latrine pits containing colour-marked faeces, the distance between these bait returns and their original setts, and patterns of overlap between mapped home ranges were all related to the culling treatments applied. Importantly, and in contrast with previous studies, consistent differences between treatments were detected across multiple study areas with substantially different culling histories, indicating that they probably represent general patterns.

The number of latrine pits found to contain colour-marked faeces (bait returns) is not usually used as a measure of badger abundance (Delahay et al. 2000). However, both the number of latrines encountered away from setts (Tuyttens et al. 2001) and the number of faecal deposits encountered at setts in spring (Wilson et al. 2003) have been found to correlate with local badger density, suggesting that the numbers of bait returns might reasonably be expected to give an approximate index of badger numbers. Hutchings, Service & Harris (2002) suggested that badgers living at low densities may be less likely to defecate in latrines, potentially reducing the detectability of bait returns. This offers an alternative mechanism whereby the number of bait returns per sett might be lower at lower badger densities, but does not undermine the plausibility of a link between bait returns and badger abundance.

If the number of bait returns per sett does give an approximate index of badger numbers, then our findings indicate that, as expected, badger densities are suppressed by culling. This effect was greatest in proactive areas, but Fig. 2a shows that bait returns were also markedly lower in reactive areas than in areas not subjected to experimental culling. This argues against the hypothesis that the apparently negative effects of reactive culling on cattle TB might have been caused by widespread illegal culling of badgers in experimental control areas (Godfray et al. 2004). While these data indicate that reactive culling influenced badger activity over a scale of several square kilometres, analyses revealed no effect of particular setts’ exposure to reactive culling at a more local scale. Failure to detect such an effect could result from the low power of the analyses performed on data from within reactive areas, but could also reflect changes in the distribution of badger activity post-cull, dissipating localized effects over larger areas.

Bait returns suggest a reduced level of badger activity inside proactive culling areas, but badger populations appear to have persisted nevertheless, with a proportion of setts showing high levels of activity (Fig. 1). The persistence of highly active setts deep inside proactive culling areas relates to a variety of factors, including lack of landowner consent for culling and disruption of culls by animal rights activists.

While the number of bait returns per sett probably gives only an approximate index of local badger density, this measure suggests that culling may influence badger populations beyond the area actually culled. Overall, the average number of bait returns recovered per sett outside proactive areas was comparable with that recorded in no-culling areas (Fig. 2a). However, this measure declined with increasing proximity to the boundary of the culling area (Fig. 3a). Taken together, these results suggest that, in contrast with findings from more wide-ranging species (Woodroffe & Frank 2005), culling probably does not have a major impact upon the density of badgers in adjoining areas (consistent with Kruuk & Macdonald's 1985 portrayal of badgers as a ‘contractionist’ species that rarely expands its home ranges to occupy vacant space) but does have a small, measurable, effect on badger density and possibly an important effect on local population dynamics.

The distances moved by badgers, measured as the median distance between a bait return and its sett of origin, also varied in response to culling (Fig. 2b). While the spatial distribution of bait returns has been shown to give a good approximation to the location of territory boundaries in high-density populations (Kruuk 1978; Delahay et al. 2000), it is known to be a less reliable indicator of badger spatial organization in low-density populations (Kruuk & Parish 1982; Delahay et al. 2000). It is important to bear in mind that badgers use defecation for scent marking not just for excretion. Hence, while the distribution of bait returns might change (or not) in response to culling-induced changes in territorial organization, this does not necessarily demonstrate that movement patterns have (or have not) changed. If badgers move through an area, but do not defecate there, those movements will not be detected by this method. Interestingly, Tuyttens et al. (2000) found temporal changes in the distribution of bait returns that they attributed to the effects of culling, but could detect no changes in the movements of individual radio-collared badgers.

Despite these caveats, the evidence strongly suggests that badgers’ spatial organization was influenced by culling. Hence, our results uphold the findings of previous, less extensive, studies showing that culling prompts wide-ranging behaviour (Cheeseman et al. 1993; O’Corry-Crowe et al. 1996; Tuyttens et al. 2000). Moreover, the data presented in Fig. 3b suggest that the effects of culling on badger movement patterns might not be confined to the area actually culled. Outside proactive culling areas, the median distance to bait returns was greatest for setts in sectors closest to the boundary of the culling area. Hence, any consequences of badger movement patterns for TB transmission might also extend beyond the boundaries of the culling area.

The probability of contact between badgers, and between badgers and cattle, will be influenced by both the density of badgers and the extent of their movements. Bait returns suggested that, inside proactive culling areas, badger abundance was markedly reduced (suggesting lower probabilities of contact) but ranging behaviour increased (suggesting higher probabilities of contact). Not surprisingly, badger abundance appears to have been reduced to a lesser extent in reactive culling areas, but ranging behaviour was also somewhat elevated. Whether the combined effects of reduced density and more extensive movements could generate badger–cattle contact probabilities higher than those occurring in survey-only areas depends upon whether badger density or movement has the greatest influence on contact rates; this is unknown, although simple models might help to characterize the patterns. Nevertheless, these patterns do provide a possible biological mechanism that may explain the apparently greater incidence of cattle TB in reactive culling areas, despite reduced badger density. The plausibility of this argument depends in part on the time it would take for badger culling to generate additional cases of TB in cattle (Godfray et al. 2004). Behavioural data show that local reductions in badger density cause badgers to alter their ranging behaviour within a few days or weeks (Cheeseman et al. 1993; Roper & Lüps 1993; Woodroffe et al. 1995). suggesting that badger-cattle contact rates would change rapidly after culling. The time taken for these contacts to lead to new infections in cattle is unknown but presumably variable. Following infection, cattle become responsive to the tuberculin test after about three weeks (C. Howard, Institute for Animal Health, personal communication, cited in Le Fevre et al. (2005)). Hence, if badgers can infect susceptible cattle rapidly on contact, increased cattle incidence would be detectable 2–3 months after badger culling. Alternatively (or additionally), expanded badger movements might influence cattle TB incidence through greater transmission among badgers, with consequently higher prevalence, ultimately causing greater transmission to cattle (Swinton et al. 1997; Tuyttens et al. 2000). In this scenario, however, new infections in cattle would appear more slowly because an additional (badger-badger) transmission stage would be involved. To summarise, the patterns we observed could plausibly explain the apparently greater incidence of cattle TB in reactive culling areas, on the timescale at which it was detected. However, as reactive culling selectively removes badgers that are spatially associated with infected cattle and hence more likely to be infected themselves (Woodroffe et al. 2005), rates of contact with infectious badgers might not follow the same pattern as contacts with badgers in general. Hence, while our findings are consistent with the hypothesis that reactive culling may have influenced TB risks to cattle through perturbation of badger spatial organization, they do not provide a conclusive test of this hypothesis.

If localized culling does indeed influence contact rates between badgers and cattle in the manner that we hypothesize, this may have implications for the effectiveness of culling carried out on a larger spatial scale, as in the proactive treatment of the RBCT and in a similar study recently completed in Ireland (Griffin et al. 2005). Inside proactive culling areas, badger densities are substantially reduced, potentially countering effects of increased movement on contact rates with cattle. Hence, detrimental effects of badger culling on cattle TB incidence could be smaller than in reactive areas, or even non-existent. However, on the edges of proactive culling areas, including adjoining areas not subjected to culling, badger densities are less depressed but movements do appear to be increased. Therefore, if contact rates with badgers influence TB risks for cattle, these areas might be expected to experience elevated risks.

In conclusion, our results suggest that culling can profoundly affect both the density and the ranging behaviour of badgers, and this has possible implications for the transmission of M. bovis to cattle. Bait-marking data provide no support for the suggestion that illegal culling in experimental control areas might explain why the incidence of cattle TB is lower in these areas than in nearby areas subjected to localized badger culling. The data do, however, indicate that badgers in and around areas subject to culling range more widely than those in undisturbed populations, potentially increasing their contact rates both with cattle and other badgers. This is consistent with the observation that culling strategies that remove only a small proportion of local badger populations, such as the ‘interim strategy’ that operated from 1986 to 1998, and the reactive culling treatment of the RBCT, appear either ineffective or counter-productive (Krebs et al. 1997; Donnelly et al. 2003; Le Fevre et al. 2005). These findings may help to design more effective management policies, and should be taken into account in determining what role badger culling should play in future strategies to control cattle TB.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

All of the data presented here were collected by staff of the Department for Environment, Food and Rural Affairs's Wildlife Unit, following training by staff of the Central Science Laboratory; we are extremely grateful for the hard work by all involved. We also wish to thank the many land owners and occupiers for consent to work on their land.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Barlow, N.D. (1996) The ecology of wildlife disease control: simple models revisited. Journal of Applied Ecology, 33, 303314.
  • Bourne, J., Donnelly, C., Cox, D., Gettinby, G., McInerney, J.P., 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. Ministry of Agriculture, Fisheries & Food, London, UK.
  • Bourne, J., Donnelly, C.A., Cox, D.R., Gettinby, G., McInerney, J.P., Morrison, W.I. & Woodroffe, R. (1999) An Epidemiological Investigation Into Bovine Tuberculosis: Towards a Sustainable Policy to Control TB in Cattle. Second Report of the Independent Scientific Group on Cattle TB. Ministry of Agriculture, Fisheries & Food, London, UK.
  • Bourne, J., Donnelly, C.A., Cox, D.R., Gettinby, G., McInerney, J.P., Morrison, W.I. & Woodroffe, R. (2004) Response of the Independent Scientific Group on Cattle TB (ISG) to the Independent Scientific Review of the Randomised Badger Culling Trial and Associated Epidemiological Research. HMSO, London, UK.
  • Bourne, J., Donnelly, C.A., Cox, D.R., Gettinby, G., McInerney, J.P., Morrison, W.I. & Woodroffe, R. (2005) An Epidemiological Investigation Into Bovine Tuberculosis: Towards a Science-Based Control Strategy. Fourth Report of the Independent Scientific Group on Cattle TB. Department for Environment, Food & Rural Affairs, London, UK.
  • British Veterinary Association (2005) Supporting document for BVA's bovine tuberculosis (bTB) policy statement, strategy document and list of recommendations. http://www.bva.co.uk/policy/issues/btb_supporting_doc.pdf
  • Cheeseman, C.L., Mallinson, P.J., Ryan, J. & Wilesmith, J.W. (1993) Recolonisation by badgers in Gloucestershire. The Badger (ed. T.J.Hayden), pp. 7893. Royal Irish Academy, Dublin, Ireland.
  • Clifton-Hadley, R.S., Wilesmith, J.W., Richards, M.S., Upton, P. & Johnston, S. (1995) The occurrence of Mycobacterium bovis infection in and around an area subject to extensive badger (Meles meles) control. Epidemiology and Infection, 114, 179193.
  • 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 organization of the European badger (Meles meles). Mammal Review, 30, 7387.
  • Donnelly, C.A., Woodroffe, R., Cox, D.R., Bourne, J., Gettinby, G., Le Fevre, A.M., McInerney, J.P. & Morrison, W.I. (2003) Impact of localized badger culling on TB incidence in British cattle. Nature, 426, 834837.
  • Dunnet, G.M., Jones, D.M. & McInerney, J.P. (1986) Badgers and Bovine Tuberculosis: Review of Policy. HMSO, London, UK.
  • Eves, J.A. (1999) Impact of badger removal on bovine tuberculosis in east County Offaly. Irish Veterinary Journal, 52, 199203.
  • Gallagher, J. et al. (2005) The Control of Bovine Tuberculosis: a Vote of No Confidence. http://www.owenpaterson.org.uk/record.jsp?type=news&ID=83 .
  • Godfray, H.C.J., Curnow, R.N., Dye, C., Pfeiffer, D., Sutherland, W.J. & Woolhouse, M.E.J. (2004) Independent Scientific Review of the Randomised Badger Culling Trial and Associated Epidemiological Research. HMSO, London, UK.
  • Griffin, J.M., Williams, D.H., Kelly, G.E., Clegg, T.A., O'Boyle, I., Collins, J.D. & More, S.J. (2005) The impact of badger removal on the control of tuberculosis in cattle herds in Ireland. Preventive Veterinary Medicine, 67, 237266.
  • Hutchings, M.R., Service, K.M. & Harris, S. (2002) Is population density correlated with faecal and urine scent marking in European badgers (Meles meles) in the UK? Mammalian Biology, 67, 286293.
  • Krebs, J.R., Anderson, R., Clutton-Brock, T., Morrison, I., Young, D., Donnelly, C., Frost, S. & Woodroffe, R. (1997) Bovine Tuberculosis in Cattle and Badgers. HMSO, London, UK.
  • Kruuk, H. (1978) Spatial organization and territorial behaviour of the European badger Meles meles. Journal of Zoology, 184, 119.
  • Kruuk, H. & Macdonald, D.W. (1985) Group territories of carnivores: empires and enclaves. Behavioural Ecology: Ecological Consequences of Adaptive Behaviour (eds R.M.Sibly & R.H.Smith). Blackwell Scientific Publications, Oxford, UK.
  • Kruuk, H. & Parish, T. (1982) Factors affecting population density, group size and territory size of the European badger, Meles meles. Journal of Zoology, 196, 3139.
  • 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/animalh/tb/isg/pdf/lefevre1005.pdf.
  • Morris, R.S., Pfeiffer, D.U. & Jackson, R. (1994) The epidemiology of Mycobacterium bovis infections. Veterinary Microbiology, 40, 153177.
  • National Farmers’ Union (2005) NFU Proposals to DEFRA on Badger Control: Disease Reduction through Sustainable Wildlife Management. National Farmers’ Union, London, UK.
  • Neal, E. & Cheeseman, C. (1996) Badgers. Poyser, London, UK.
  • 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, 147158.
  • Rogers, L.M., Delahay, R.J., 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. Proceedings of the Royal Society of London, Series B, 265, 18.
  • Roger, T.J. & Lüps, P. (1993) Disruption of territorial behaviour in badgers Meles meles. Zeitschrift für Saugetierkunde, 58, 252255.
  • Roper, T.J., Sheperdson, D.J. & Davies, J.M. (1986) Scent marking with faeces and anal secretion in the European badger (Meles meles): seasonal and spatial characteristics of latrine use in relation to territoriality. Behaviour, 97, 94117.
  • Swinton, J., Tuyttens, F., Macdonald, D.W., Nokes, D.J., Cheeseman, C.L. & Clifton-Hadley, R. (1997) A comparison of fertility control and lethal control of bovine tuberculosis in badgers: the impact of perturbation induced transmission. Philosophical Transactions of the Royal Society, 352, 619631.
  • Tuyttens, F.A.M., Delahay, R.J., Macdonald, D.W., Cheeseman, C.L., Long, B. & Donnelly, C.A. (2000) Spatial perturbation caused by a badger (Meles meles) culling operation: implications for the function of territoriality and the control of bovine tuberculosis (Mycobacterium bovis). Journal of Animal Ecology, 69, 815828.
  • Tuyttens, F.A.M., Long, B., Fawcett, T., Skinner, A., Brown, J.A., Cheeseman, C.L. & Roddam, A.W. & Macdonald, D.W. (2001) Estimating group size and population density of Eurasian badgers Meles meles by quantifying latrine use. Journal of Applied Ecology, 38, 11141121.
  • Tuyttens, F.A.M. & Macdonald, D.W. (2000) Consequences of social perturbation for wildlife management and conservation. Behaviour and Conservation (edsL.M.Gosling & W.J.Sutherland), pp. 315329. Cambridge University Press, Cambridge.
  • White, P.C.L. & Harris, S. (1995) Bovine tuberculosis in badger (Meles meles) populations in southwest England: an assessment of past, present and possible future control strategies using simulation modelling. Philosophical Transactions of the Royal Society of London, Series B, 349, 415432.
  • Wilson, G.J., Delahay, R.J., de Leeuw, A.N.S., Spyvee, P.D. & Handoll, D. (2003) Quantification of badger (Meles meles) sett activity as a method of predicting badger numbers. Journal of Zoology, 259, 4956.
  • Woodroffe, R. & Frank, L.G. (2005) Lethal control of African lions (Panthera leo): local and regional population impacts. Animal Conservation, 8, 9198.
  • Woodroffe, R. & Macdonald, D.W. (1993) Badger sociality: models of spatial grouping. Symposia of the Zoological Society of London, 65, 145169.
  • Woodroffe, R., Bourne, F.J., Donnelly, C.A., Cox, D.R., Gettinby, G., McInerney, J.P. & Morrison, W.I. (2002) Towards a sustainable policy to control TB in cattle. Conservation and Conflict: Mammals and Farming in Britain (eds F.Tattersall & W.Manley). Linnean Society, London, UK.
  • Woodroffe, R., Donnelly, C.A., Johnston, W.T., Bourne, F.J., Cheeseman, C.L., Clifton-Hadley, R.S., Cox, D.R., Gettinby, G., Hewinson, R.G., Le Fevre, A.M., McInerney, J.P. & Morrison, W.I. (2005) Spatial association of Mycobacterium bovis infection in cattle and badgers. Journal of Applied Ecology, 42, 852862.
  • Woodroffe, R., Macdonald, D.W. & da Silva, J. (1995) Dispersal and philopatry in the European badger, Meles meles. Journal of Zoology, 237, 227239.
  • Zuckerman, S. (1980) Badgers, Cattle and Tuberculosis. HMSO, London, UK.