SEARCH

SEARCH BY CITATION

Keywords:

  • bovine tuberculosis;
  • connectivity;
  • culling;
  • dispersal;
  • least-cost modelling;
  • vaccination

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Landscape epidemiological studies often produce maps that visualise spatial variation in disease incidence or prevalence. Within England and Wales a component of the spatial variation in the incidence of bovine tuberculosis (bTB) is endemic infection in the European badger (Meles meles) population. This reservoir of infection presents a major obstacle to control and eradication of the disease in cattle, which is currently one of the major issues in agricultural policy in England and Wales. Previous bTB maps have considered the distribution and abundance of badgers, but there is recognition in epidemiology that other landscape characteristics that may relate to disease risk need to be considered. Landscape isolation is particularly relevant for management of disease in wildlife hosts, and especially for bTB in badgers, as the potential for badgers to disperse into and out of areas under management affects the success of culling, and possibly also the success of vaccination. To map the relative levels of landscape isolation for badgers, we used expert opinion of landscape obstacles to dispersal to apply a continuous surface catchment area approach, which is based on the concept of landscape connectivity and the technique of least-cost modelling. Our results indicate that some parts of the landscape are relatively more or less isolated than others. This finding is directly relevant in the context of potential bTB management strategies, and we hope this will encourage collation of data designed to quantify the degree of landscape isolation for wildlife disease hosts, and encourage its consideration when formulating and assessing disease management strategies.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Landscape, or spatial, epidemiology is a scientific discipline that uses geographical principles, techniques and tools to understand the spatial variation of disease risk or incidence (Lambin et al. 2010; Ostfeld et al. 2005). An example of a zoonotic disease whose management could benefit from a landscape epidemiology perspective is that of bovine tuberculosis (bTB). In the UK and the Republic of Ireland, endemic infection of bTB in the European badger (Meles meles) population is a major obstacle to the control and eventual eradication of the disease in cattle (Gortázar et al. 2012). In England and Wales, the control of bTB is currently one of the major issues in agricultural policy, as the social and economic costs associated with incidence of the disease in cattle are considered unsustainable by both countries' governments (Defra 2012; Welsh Government 2012).

A key output of landscape epidemiology that emphasises its geographical nature is the production of maps that visualise disease incidence and risk across a landscape (Ostfeld et al. 2005). Such risk maps have been produced to examine geographical differences in the risk of transmission of bTB from badgers and deer to cattle within England and Wales (Ward et al. 2009; White et al. 2008). However, these previous efforts have been based solely on the distribution of wildlife reservoirs of the disease, and there is a recognised need to include other factors such as landscape connectivity in order better to understand disease risk, spread and containment (Lambin et al. 2010; Ostfeld et al. 2005).

Landscape connectivity can be defined as the degree to which a landscape facilitates or impedes movements (Taylor et al. 1993; Tischendorf and Fahrig 2000), and will affect processes such as dispersal (Wiens 2001). This dependence of dispersal on landscape connectivity will result in a geography of isolation, in which some locations within the landscape are harder to disperse to or from and can therefore be considered more isolated. A local wildlife population with greater isolation may have reduced migration rates, and therefore may exhibit simpler disease dynamics that make understanding and managing the disease within the local population easier (Delahay et al. 2009).

This notion of landscape isolation and its limits on dispersal has particular relevance to badger culling, which is the main strategy that has been used for trying to reduce the risk of bTB transmission from badgers to cattle (Krebs et al. 1997). In high-density, undisturbed badger populations, dispersal tends to be limited both in terms of incidence of dispersal events and distances (Cheeseman et al. 1988; Roper et al. 2003; Woodroffe et al. 1993). However, culling has been shown to cause social perturbation of the badger population, which results in an increase in badger movement and migration (Carter et al. 2007; Woodroffe et al. 2006). This social perturbation is thought to be the reason that while a beneficial effect on bTB incidence in cattle can be achieved within a badger culling area, there is an associated, temporary detrimental impact on cattle disease incidence in surrounding areas (Donnelly et al. 2006; Jenkins et al. 2010). Rates of movement and changes in social group size in undisturbed populations have also been shown to be associated with increased disease incidence in badgers (Rogers et al. 1998; Vicente et al. 2007). Therefore, management strategies involving culling may achieve greater overall benefits in landscapes that are more isolated from the perspective of the badger population, as the effects from social perturbation could be reduced.

A recently licensed injectable Bacillus Calmette-Guérin vaccine for badgers significantly reduces the severity and progression of infection in vaccinated badgers (Chambers et al. 2011) and the risk of infection in unvaccinated cubs within vaccinated social groups (Carter et al. 2012). Therefore, vaccination of badgers could comprise a component of a programme of measures aimed at controlling bTB in cattle (Smith et al. 2012; Wilson et al. 2011). However, the feasibility and success of a vaccination strategy may also be affected in part by migration rates of individuals within the vaccinated population (Delahay et al. 2003). Therefore, although the significance of landscape isolation for controlling bTB within a local badger population via vaccination is not as clear as for culling, it remains relevant.

Given the likely effects of isolation on either culling or vaccination approaches to managing bTB within the badger population, precedents for considering the isolating effects of geography in the management of bTB in badgers do exist. For instance, when establishing culling trials in the Republic of Ireland (Griffin et al. 2005) and England (Donnelly et al. 2006), study areas were designed where possible to be bounded by landscape features such as rivers and coastlines that were expected to be barriers to badger movement. However, for isolation to be properly considered within a landscape epidemiological approach to bTB risk assessment or management planning, isolation needs to be quantified and visualised across the entire landscape. Therefore, the objective of this work is to use an expert opinion-based modelling approach to map isolation for badgers across England and Wales.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Badger catchment areas

Landscape isolation for badgers was modelled using a continuous surface catchment area approach (Etherington 2012). We consider this to be an intuitive approach to establishing how isolated a location is in its surrounding landscape, as catchments are used throughout geography for measuring processes such as flows and accessibility. In this context, a badger catchment around a location represents the landscape to or from which badger dispersal is likely to occur. Locations with smaller badger catchment areas are considered more isolated in comparison to locations with larger badger catchment areas. It is important to note that in adopting this catchment area approach we assume that the landscape is the primary factor limiting badger dispersal.

The continuous surface catchment area approach (Etherington 2012) is based on least-cost modelling (Adriaensen et al. 2003; Douglas 1994). The key component of least-cost modelling is a geographic information system (GIS) cost-surface. A cost-surface is a raster map in which each cell has a value that, in ecological terms, describes the functional connectivity (Tischendorf and Fahrig 2000) of that cell as a traversal cost. This cost is some function of behavioural aversion, energetic expenditure and mortality risk to movement for a species given the landscape represented by the raster cell. Given a source cell on the cost-surface, least-cost modelling produces an accumulated-cost-surface that gives the least-cost path between the source cell and all other cells as a function of the distance travelled and the costs traversed. By specifying a maximum accumulated-cost that represents the likely limit of dispersal for a species, a catchment can be delineated and its area calculated. By repeating this process by using all landscape cells as a source cell, a catchment area can be calculated for all cells in order to produce a catchment area map.

The two fundamental data requirements needed to model landscape isolation using a continuous surface catchment area approach are: a cost-surface that represents the dispersal costs across a landscape, and a maximum accumulated-cost that represents the likely limit of dispersal. Unfortunately, neither of these data requirements exists for badgers. No cost-surface has ever been developed, and dispersal has only ever been reported in terms of distances rather than accumulated-costs. This means that existing records of badger dispersal distances can only be used as a guide in choosing an appropriate accumulated-cost value as they do incorporate the effects of landscape on dispersal that are being modelled. At landscape scales such ecological data gaps are quite common, and this is why we will use expert opinion to supplement and interpret the available empirical data (Perera et al. 2012).

Landscape obstacle occurrence data

The first decision to make in creating a cost-surface is to choose an appropriate resolution. As the catchment area approach is computationally demanding, the optimal raster resolution should be as coarse as possible while remaining ecological meaningful (Etherington 2012). As badger dispersal is defined in terms of a permanent movement between badger social group territories (Cheeseman et al. 1988; Roper et al. 2003; Woodroffe et al. 1993), we chose a cell resolution of 500 m because this created cells with an area of 25 ha, which approximates the smallest size of badger territory that would be expected to occur commonly in rural England and Wales (Neal and Cheeseman 1996).

In terms of choosing what geographic data are important to include in creating a cost-surface, no definitive information exists on what landscape features are most important for limiting badger dispersal. An analysis of road mortality data indicated that the effect of roads would vary depending on the size of a road and its traffic levels, and will be a function of direct mortality, the presence of fencing and behavioural aversion (Clarke et al. 1998). Use of genetic data found that motorways present at least some form of obstacle to gene flow, and that rivers at least 50 m wide present a significant obstacle (Frantz et al. 2010). In culled areas, badger immigration, as measured through continued trapping, showed a minimal effect from rivers around 10 m wide, but a significant effect from open water (Sleeman et al. 2009). Given this limited information, we decided to follow other models of badger dispersal (Schippers et al. 1996; Van Apeldoorn et al. 1998) by assuming that obstacles would include poor-quality habitat, roads and rivers. In order to incorporate each of these landscape obstacles into a cost-surface, GIS data for each landscape obstacle was used to create a raster that represents the occurrence of the landscape obstacle on a scale from zero to one.

The obstacle effect of poor-quality habitat was based on a raster dataset that describes habitat quality for badgers in terms of the likely density of badger main setts across England and Wales (Etherington et al. 2009). The inverse of such habitat suitability models can be easily incorporated into a cost-surface, such that poorer habitat has greater cost and therefore acts as an obstacle to dispersal (Chetkiewicz and Boyce 2009). To achieve this, the badger main sett density model was resampled, rescaled and inverted to produce a 500 m resolution raster describing the presence of poor habitat on a continuous scale from zero to one (Figure 1a). Null data cells representing open water at least 100 m wide were treated as complete barriers to dispersal.

figure

Figure 1. Raster GIS layers of the occurrence of landscape features chosen to act as obstacles to badger dispersal in creating a cost-surface

Download figure to PowerPoint

Vector road data (Ordnance Survey 2012) and river data (Vogt et al. 2007) was rasterised with the same resolution and extent as the poor-habitat raster (Figure 1a) to identify the presence or absence, using cell values of zero and one respectively, of: trunk A-roads (Figure 1b), motorways (Figure 1c), medium rivers (Figure 1c) and large rivers (Figure 1d). Care was taken during rasterisation of the vector linear obstacle data to ensure no gaps were introduced (Adriaensen et al. 2003).

Landscape dispersal scenarios

The landscape obstacle occurrence rasters (Figure 1) were combined into a cost-surface using a map algebra point operation (Equation (1)). A cost-surface (cs) was created from the set (S) of landscape obstacle occurrence rasters by multiplying each raster (r) by an associated weight (wr) that describes that raster's relative importance as an obstacle to dispersal.

  • display math(1)

As with choosing what landscape features are most important in regards to limiting badger dispersal, there is no empirical data available to guide the values chosen for the weights used in creating the cost-surface, or indeed the maximum accumulated-cost that represents the likely limit of badger dispersal that is required to generate catchments from a cost-surface. This is a common dilemma in applying least-cost modelling, a solution for which is to use expert opinion to establish a set of feasible scenarios (Zeller et al. 2012). We used a panel of nine badger experts drawn from within the Animal Health and Veterinary Laboratories Agency. All the experts had published scientific work involving badger movements, and between them had experience of studying badger populations in a range of locations, densities and situations at a variety of research institutions. Using their knowledge of the scientific literature and their own personal field experience, each expert was independently interviewed in order to estimate a set of weightings that represented their perception of the relative costs of each landscape obstacle to badger dispersal, and a maximum accumulated-cost to represent a likely limit to the dispersal ability of badgers. As with choice of modelling resolution, the computational demands of the catchment area technique restricted the number of scenarios that could be examined. Therefore we analysed 12 different landscape dispersal scenarios that best reflected the distribution of estimated values provided by the experts (Table 1). We first defined scenarios in which the river and road obstacles were weighted by values that were at the bottom, middle and top of the range of values suggested by the experts. Each of these three sets of river and road weights were then applied with each combination of the poor-habitat weight and maximum accumulated-cost values that were the bottom and top of the ranges suggested by the experts.

Table 1. The combinations of weightings for landscape obstacles and values for maximum accumulated-cost used for each of the 12 landscape dispersal scenarios
Landscape dispersal scenarioLandscape obstacle cost-surface weightingMaximum accumulated-cost
MotorwayTrunk A-roadLarge riverMedium riverPoor habitat
 15297110 000
 2529715 000
 35297210 000
 4529725 000
 531.58.56110 000
 631.58.5615 000
 731.58.56210 000
 831.58.5625 000
 972.59.58110 000
1072.59.5815 000
1172.59.58210 000
1272.59.5825 000

Relative isolation maps

As our modelling was based on expert opinion rather than empirical data, we felt it was more appropriate to present our isolation estimates as an index. Therefore, we chose to rescale each catchment area map from zero, indicating the largest catchment area and hence the least isolated part of the landscape, to one, indicating the smallest catchment area and hence the most isolated part of the landscape. Therefore the results are presented in terms of relative isolation within a landscape.

When applying a catchment area approach it is important to consider map edge effects (Etherington 2012). For example, it is important to remember that badgers in northern England may well migrate across the border with Scotland. Therefore to ensure that all catchment areas are not artificially reduced by spatial limits in the extent of the data used, areas within 10 km of the England–Scotland border were removed from the relative isolation maps. A threshold of 10 km was used because the maximum accumulated-cost value among the landscape dispersal scenarios was 10 000 (Table 1), which equates to a distance of 10 km in the absence of any obstacles.

Finally, to visualise differences between the 12 landscape dispersal scenarios, we used map algebra point operations to calculate the mean and standard deviation of the 12 relative isolation maps because we considered all 12 of the landscape dispersal scenarios to be equally plausible. The relative isolation maps were calculated using graph theory-based Python scripts (Etherington 2012), with ArcGIS version 10 (ESRI, Redlands, California, USA) used for pre- and post-processing of all GIS data.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In all of the 12 relative landscape isolation maps (Figure 2) calculated for each of the landscape dispersal scenarios (Table 1), relative isolation varied across the landscape of England and Wales. When considering the whole landscape, relative isolation was in general higher for landscape dispersal scenarios in which poor habitat was considered a greater obstacle to movement (for example, Figure 2a vs Figure 2c). When considering localised parts of the landscape, the isolating effect of linear obstacles was more prominent for scenarios in which the maximum accumulated-cost limiting badger dispersal was smaller (for example, Figure 2a vs Figure 2b). Whenever an isolating effect from linear obstacles was present, it was clearly localised to the immediate vicinity of the road or river.

figure

Figure 2. Relative isolation maps for badgers in England and Wales for each of the 12 different landscape dispersal scenarios (Table 1)

Download figure to PowerPoint

The overall spatial pattern of the relative isolation maps, in terms of which parts of the landscape were relatively more or less isolated, was consistent among the landscape dispersal scenarios (Figure 2). The maps of mean (Figure 3a) and standard deviation (Figure 3b) of the relative isolation demonstrate the consistency between the results from each of the landscape dispersal scenarios. The spatial pattern of mean relative isolation remained generally the same as in each of the individual scenario relative isolation maps. Smaller standard deviation values occurred in areas that were further from obstacles with greater uncertainty, such as poor habitat and motorways, and closer to obstacles with greater certainty such as large rivers and the absolute barrier imposed by the coastline.

figure

Figure 3. Maps of (a) the mean and (b) the standard deviation of relative landscape isolation for badgers across England and Wales from the 12 landscape dispersal scenarios

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The measures of relative isolation varied spatially within a scenario (Figure 2). This indicates that the catchment area approach used to measure isolation was able to use landscape composition to produce estimates of isolation that differentiate locations across England and Wales. The measures of relative isolation also varied between scenarios (Figure 2), which indicates that the catchment area approach was sensitive to the different model parameters used for each landscape dispersal scenario (Table 1), therefore justifying the need to compare multiple dispersal scenarios. While there was variation in relative landscape isolation between the different scenarios, the results were largely consistent, with the areas of relatively high isolation being spatially consistent across scenarios. This consistency of results suggests that parts of England and Wales are relatively more or less isolated than others.

The outcomes of the two main management options for controlling bTB in badgers, culling and vaccination, are likely to be affected by the level of landscape isolation with respect to badger dispersal, which has been shown here to vary geographically. This has implications for the design of management strategies for the control of bTB, where the degree of landscape isolation for badgers would be considered an important factor when determining how and where to deploy interventions. For example, assuming a vaccination approach was to be used, its success would depend on maintaining a high proportion of immunised badgers, which might be more difficult in less isolated areas due to higher badger migration rates. With knowledge of differences in isolation across a landscape, greater resources could be directed into less isolated areas in order to maximise the efficiency of a vaccination programme.

There are significant uncertainties associated with the use of expert opinion-based cost-surfaces (Zeller et al. 2012). Therefore, before information on the geography of badger isolation can form a reliable part of such a geographically customised bTB management strategy in England and Wales, we would recommend that the expert opinion maps of relative badger landscape isolation presented here would need to be updated to include empirical data on badger dispersal that accounts for the effects of landscape. This would enable maps to be produced describing the geography of badger landscape isolation in absolute units, which would be of more value for management than the relative units presented here. Such an isolation model would require data to inform a cost-surface, representing landscape connectivity, and a maximum accumulated-cost, representing the likely limit of badger dispersal. There are a variety of quantitative approaches to establishing cost-surfaces that are based on methods such as live-trapping, radio-tracking and landscape genetic data (Zeller et al. 2012). Live-trapping or ideally radio-tracking data on badger dispersal, particular the rarer long-distance dispersals, would also be required in order to allow the measurement of badger dispersal in accumulated-cost units rather than straight-line distance. Much of this kind of data has already been collected for badgers within England and Wales, so it should be possible to collate the required data, especially if such an effort involves multiple institutions.

To conclude, we believe that our work demonstrates that it is possible to quantify landscape isolation for badgers in England and Wales, which is likely to be relevant in the context of potential bTB management strategies. By highlighting the relevance of landscape isolation we hope to encourage collation of empirical data designed to quantify the degree of landscape isolation for badgers, and to encourage the consideration of isolation when reviewing bTB management strategies, and in landscape epidemiology risk assessments more generally.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the Welsh Government and the Department for Environment, Food and Rural Affairs for funding.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Adriaensen F, Chardon J P, De Blust G, Swinnen E, Villalba S, Gulinck H and Matthysen E 2003 The application of ‘least-cost’ modelling as a functional landscape model Landscape and Urban Planning 64 233247
  • Carter S P, Delahay R J, Smith G C, Macdonald D W, Riordan P, Etherington T R, Pimley E R, Walker N J and Cheeseman C L 2007 Culling-induced social perturbation in Eurasian badgers Meles meles and the management of TB in cattle: an analysis of a critical problem in applied ecology Proceedings of the Royal Society B 274 27692777
  • Carter S P, Chambers M A, Rushton S P, Shirley M D F, Schuchert P, Pietravalle S, Murray A, Rogers F, Gettinby G, Smith G C, Delahay R J, Hewinson R G and McDonald R A 2012 BCG vaccination reduces risk of tuberculosis infection in vaccinated badgers and unvaccinated badger cubs PLoS ONE 7 18
  • Chambers M A, Rogers F, Delahay R J, Lesellier S, Ashford R, Dalley D, Gowtage S, Davé D, Palmer S, Brewer J, Crawshaw T, Clifton-Hadley R, Carter S, Cheeseman C, Hanks C, Murray A, Palphramand K, Pietravalle S, Smith G C, Tomlinson A, Walker N J, Wilson G J, Corner L A L, Rushton S P, Shirley M D F, Gettinby G, McDonald R A and Hewinson R G 2011 Bacillus Calmette-Guerin vaccination reduces the severity and progression of tuberculosis in badgers Proceedings of the Royal Society B 278 19131920
  • Cheeseman C L, Cresswell W J, Harris S and Mallinson P J 1988 Comparison of dispersal and other movements in two badger (Meles meles) populations Mammal Review 18 5159
  • Chetkiewicz C L B and Boyce M S 2009 Use of resource selection functions to identify conservation corridors Journal of Applied Ecology 46 10361047
  • Clarke G P, White P C L and Harris S 1998 Effects of roads on badger Meles meles populations in south-west England Biological Conservation 86 117124
  • Delahay R J, Wilson G J, Smith G C and Cheeseman C L 2003 Vaccinating badgers (Meles meles) against Mycobacterium bovis: the ecological considerations The Veterinary Journal 166 4351
  • Delahay R J , Smith G C and Hutchings M R eds 2009 Management of disease in wild mammals Springer, Tokyo
  • Department for Environment Food and Rural Affairs [Defra] 2012 Bovine TB (http://www.defra.gov.uk/animal-diseases/a-z/bovine-tb/) Accessed 11 December 2012
  • Donnelly C A, Woodroffe R, Cox D R, Bourne F J, Cheeseman C L, Clifton-Hadley R S, Wei G, Gettinby G, Gilks P, Jenkins H, Johnston W T, Le Fevre A M, McInerney J P and Morrison W I 2006 Positive and negative effects of widespread badger culling on tuberculosis in cattle Nature 439 843846
  • Douglas D H 1994 Least-cost path in GIS using an accumulated cost surface and slopelines Cartographica 31 3751
  • Etherington T R 2012 Mapping organism spread potential by integrating dispersal and transportation processes using graph theory and catchment areas International Journal of Geographical Information Science 26 541556
  • Etherington T R, Ward A I, Smith G C, Pietravalle S and Wilson G J 2009 Using the Mahalanobis distance statistic with unplanned presence-only survey data for biogeographical models of species distribution and abundance: a case study of badger setts Journal of Biogeography 36 845853
  • Frantz A C, Pope L C, Etherington T R, Wilson G J and Burke T 2010 Using isolation-by-distance-based approaches to assess the barrier effect of linear landscape elements on badger (Meles meles) dispersal Molecular Ecology 19 16631674
  • Gortázar C, Delahay R J, McDonald R A, Boadella M, Wilson G J, Gavier-Widen D and Acevedo P 2012 The status of tuberculosis in European wild mammals Mammal Review 42 193272
  • Griffin J M, Williams D H, Kelly G E, Clegg T A, O'Boyle I, Collins J D and More S J 2005 The impact of badger removal on the control of tuberculosis in cattle herds in Ireland Preventive Veterinary Medicine 67 237266
  • Jenkins H E, Woodroffe R and Donnelly C A 2010 The duration of the effects of repeated widespread badger culling on cattle tuberculosis following the cessation of culling PLoS ONE 5 17
  • Krebs J R, Anderson R, Clutton-Brock T, Morrison I, Young D and Donnelly C 1997 Bovine tuberculosis in cattle and badgers MAFF Publications, London
  • Lambin E F, Tran A, Vanwambeke S O, Linard C and Soti V 2010 Pathogenic landscapes: interactions between land, people, disease vectors, and their animal hosts International Journal of Health Geographics 9 113
  • Neal E G and Cheeseman C L 1996 Badgers Poyser Natural History, London
  • Ordnance Survey 2012 Meridian 2 (http://www.ordnancesurvey.co.uk/oswebsite/products/meridian2/) Accessed 26 January 2012
  • Ostfeld R S, Glass G E and Keesing F 2005 Spatial epidemiology: an emerging (or re-emerging) discipline Trends in Ecology & Evolution 20 328336
  • Perera A H , Drew C A and Johnson C J eds 2012 Expert knowledge and its application in landscape ecology Springer, New York
  • Rogers L M, Delahay R, Cheeseman C L, Langton S, Smith G C and 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 B 265 12691276
  • Roper T J, Ostler J R and Conradt L 2003 The process of dispersal in badgers Meles meles Mammal Review 33 314318
  • Schippers P, Verboom J, Knaapen J P and Van Apeldoorn R C 1996 Dispersal and habitat connectivity in complex heterogeneous landscapes: an analysis with a GIS-based random walk model Ecography 19 97106
  • Sleeman D P, Davenport J, More S J, Clegg T A, Griffin J M and O'Boyle I 2009 The effectiveness of barriers to badger Meles meles immigration in the Irish Four Area project European Journal of Wildlife Research 55 267278
  • Smith G C, McDonald R A and Wilkinson D 2012 Comparing badger (Meles meles) management strategies for reducing tuberculosis incidence in cattle PLoS ONE 7 111
  • Taylor P D, Fahrig L, Henein K and Merriam G 1993 Connectivity is a vital element of landscape structure Oikos 68 571573
  • Tischendorf L and Fahrig L 2000 On the usage and measurement of landscape connectivity Oikos 90 719
  • Van Apeldoorn R C, Knaapen J P, Schippers P, Verboom J, Van Engen H and Meeuwsen H 1998 Applying ecological knowledge in landscape planning: a simulation model as a tool to evaluate scenarios for the badger in the Netherlands Landscape and Urban Planning 41 5769
  • Vicente J, Delahay R J, Walker N J and Cheeseman C L 2007 Social organization and movement influence the incidence of bovine tuberculosis in an undisturbed high-density badger Meles meles population Journal of Applied Ecology 76 348360
  • Vogt J V, Soille P, de Jager A, inline image E, Mehl W, Haastrup P, Paracchini M L, Dusart J, Bodis K, Foisneau S and Bamps C 2007 Developing a pan-European data base of drainage networks and catchment boundaries from a 100 metre DEM in Wachowicz M and Bodum L eds 10th AGILE conference on Geographic Information Science AGILE, Aalborg 110
  • Ward A I, Smith G C, Etherington T R and Delahay R J 2009 Estimating the risk of cattle exposure to tuberculosis posed by wild deer relative to badgers in England and Wales Journal of Wildlife Diseases 45 11041120
  • Welsh Government 2012 Bovine TB (http://wales.gov.uk/topics/environmentcountryside/ahw/disease/bovinetuberculosis/?lang=en) Accessed 11 December 2012
  • White P C L, Bohm M, Marion G and Hutchings M R 2008 Control of bovine tuberculosis in British livestock: there is no ‘silver bullet’ Trends in Microbiology 16 420427
  • Wiens J A 2001 The landscape context of dispersal in Clobert J , Danchin E , Dhondt A A and Nichols J D eds Dispersal Oxford University Press, Oxford 96109
  • Wilson G J, Carter S P and Delahay R J 2011 Advances and prospects for management of TB transmission between badgers and cattle Veterinary Microbiology 151 4350
  • Woodroffe R, Macdonald D W and Dasilva J 1993 Dispersal and philopatry in the Eurasian badger Meles meles Journal of Zoology 237 227239
  • Woodroffe R, Donnelly C A, Cox D R, Bourne F J, Cheeseman C L, Delahay R J, Gettinby G, McInerney J P and Morrison W I 2006 Effects of culling on badger Meles meles spatial organization: implications for the control of bovine tuberculosis Journal of Applied Ecology 43 110
  • Zeller K A, McGarigal K and Whiteley A R 2012 Estimating landscape resistance to movement: a review Landscape Ecology 27 777797