Response to the independent audit of the badger culling trial field operations
- Executive summary
- Background
- Scope of the audit
- Appointment of auditor
- Comments on audit report
- Report recommendations
Appendix A: Independent Scientific Group on Cattle TB (ISG): response to the independent audit of the badger culling trial field operations
1. Executive summary
1.1 The randomised badger culling trial (the trial) is an important element of the Government's strategy to devise a sustainable, science based, policy to control TB in cattle. It is a large undertaking covering 3000 km2 spread across 8 counties in England and involves the participation of approximately 6,000 occupiers of land.
1.2 The audit report provides a useful vehicle for reviewing and improving two complex areas of trial work, namely, surveying for badger activity and social group territory delineation. Given the nature of the fieldwork, it was recognised from the outset that there would be imperfections in surveys of badger activity and the subsequent allocation of setts to social groups. The trial was, however, designed to test the effectiveness of culling operations carried out within the practical limits imposed by field conditions. While minimising survey errors is a priority, future policy recommendations need to be based on strategies that are practical and effective. Against this background, MAFF notes that the audit did not reveal fundamental flaws in trial standard procedures or their application. It is also noted that in its response at Appendix A, the Independent Scientific Group on Cattle TB (ISG) has confirmed that the procedures developed are being applied correctly.
1.3 MAFF has already acted in respect of many of the audit findings including those relating to greater flexibility in surveying, further training on sett classification and the use of a computer system to generate and maintain maps. Following advice from the ISG, the recommendation that setts be classified under a more complex system has not been accepted.
1.4 Given the complexity of the audit, MAFF commissioned an additional executive summary. This was submitted by the Auditors in December 2000 and has been included in the Auditors' report.
2. Background
Aims of the badger culling trial
2.1 The Krebs review found compelling evidence that badgers were a significant source of TB infection in cattle, but that the relative contributions made by badgers and other factors were unknown. It recommended a field trial to investigate the link between badgers and cattle TB. The trial, designed and overseen by the ISG, forms one element of a much wider research programme recommended by the Group. It is a vital part of the Government's work aimed at developing a long term policy, based on sound science, to control TB in cattle. The trial will address two main questions:
- what contribution badgers make to TB in cattle.
- whether culling of badgers helps to reduce TB in cattle herds.
In addition, the trial will provide information on the epidemiology of TB in cattle and badgers. This will include information on the prevalence of TB in badgers in relation to population density and social group structure and, most importantly, the spatial relationship between infected badgers and infected cattle.
Main features of the trial
2.2 Participation in the trial by landowners and occupiers is voluntary. It is being conducted in 30 areas (each of about 100 km2) where the incidence of TB in cattle is high. Trial areas are grouped into 10 'triplets' and each area within a triplet is allocated, at random, one of three treatments:
- 'proactive' culling
- where as many badgers as possible are removed (about 80%) in an initial cull and the area is kept as clear as possible for the remainder of the trial.
- 'reactive' culling
- where badgers are only removed from social groups associated with farms where TB in cattle is confirmed.
- 'survey only'
- where no badgers are trapped or killed but the land is surveyed to record badger activity.
Surveying for badger activity
2.3 Surveys of trial areas are undertaken to provide baseline estimates of badger density, to locate setts for trapping and to define the boundaries of treatment areas.
Social group territory delineation
2.4 The reactive and proactive culling treatments are designed to be applied to entire social groups of badgers. This depends on the correct allocation of setts to social groups by delineation of likely social group territories using survey data. It is recognised that in some situations clear boundaries may not be evident.
3. Scope of the audit
3.1 The auditor was instructed:
- to assess (on a sample basis) the effectiveness of surveying, social group delineation and badger culling carried out in accordance with the prescribed operating procedures; and
- in the light of findings from the previous requirement, to make recommendations for the improvement of these procedures.
3.2 The Ministry notes with regret that the potential risk to the security of the auditors arising from protest activity against the trial has led to a delay in the audit of the efficiency of trapping procedures. It is hoped that it will be the subject of a separate report which will be published as soon as possible after receipt.
3.3 In order to carry out their work effectively, the Auditors had unrestricted access to trial staff and operational documents.
4. Appointment of the auditor
4.1 Appointment of the Auditors was made through a competitive tendering process. Following an ISG assessment of the tenders submitted, Cresswell Associates were appointed in November 1998 to carry out the audit.
5. Comments on audit findings
Missed setts
5.1 The Auditors were not required to undertake resurveying from scratch but were provided with MAFF survey maps which recorded the location of setts. MAFF staff who drew up these maps were, of course, subject to time constraints and it is encouraging that despite these time limitations the Auditors found that few larger setts had been missed. The trial was designed on the basis that participation would be voluntary. It was recognised from the outset that not all land occupiers would participate and, consequently, that not all setts within proactive treatment areas could be trapped. Neither, in practice, was it realistic, given the size of trial, to expect 100% of setts to be located even where consent to survey had been given. These factors are consistent with MAFF's view that while a high trapping efficiency can be attained the trial will not result in the local eradication of badgers. Having said that, MAFF accepts that eight larger setts were missed in the area audited and that the Auditors' recommendations for improving efficiency must be taken up.
Mis-classified setts
5.2 The classification of setts in field conditions is, to some extent, subjective and, as indicated by the Auditors' recommendation for a system describing five different sett types, an often complex matter. The Auditors found that a number of fox holes and rabbit burrows were incorrectly classified as setts. In fact, in some cases, staff deliberately classified such holes as setts (providing they were large enough to accommodate a badger). This was because of their potential use by badgers during the latter stages of trapping operations when populations become highly disrupted and the use of such holes might change. Trapping at such locations would not, in any event, have led to the compromise of trial data.
Delineation of territories
5.3 The delineation of badger social group territories is a complex process. Identification of clear boundaries may not be possible where badger populations have been severely disrupted. Given these complexities, it was necessary to adapt standard procedures in the light of early feedback from field operations. The evolution of procedures was an area where advice from the ISG was of paramount importance and MAFF is particularly grateful to those Group members who helped train staff in the complex techniques involved. It is reassuring that the Auditors found the procedures developed to be sound and that the Group has recorded in its response at Appendix A that it is content that they are being applied correctly.
5.4 A particular milestone in developing delineation procedures was the introduction, early in 1999, of factors indicating the confidence with which setts could be allocated to social groups. Any delineation work undertaken prior to this time and which was reviewed in the audit would not reflect the use of confidence factors.
5.5 The delineation of territories is particularly important for the reactive strategy. As noted in the audit report, there is scope for discrepancies occurring during the original survey to be corrected in the further surveys undertaken prior to the culling of social groups associated with the breakdown farm. Delineation for the proactive treatment is less critical for trapping efficiency but MAFF notes that where original classifications were correct (over 60%) there was good agreement between boundaries delineated using trial procedures and those derived from bait-marking with 100% of setts assigned to the correct territories. Importantly, the Auditor found that confidence factors (i.e. indices used to indicate the certainty with which setts can be allocated to social groups) were effective in highlighting those setts for which assignment was unreliable. As recognised at the outset by the ISG, these points will need to be taken into account in the future analysis of data.
Bait-marking
5.6 Bait marking (i.e. a technique in which indigestible markers are fed to badgers and traced later to help delineate territories) is resource intensive and, given the seasonal variation in its value, was an option not open to MAFF staff. The Auditors were able to use this technique given the much smaller scale of the audit check. It is an indication of the complexity of the process that even with this advantage it was necessary to undertake two rounds of bait-marking in order to interpret some data.
6. Report recommendations
Missed setts
6.1 MAFF accepts that staff missed some badger setts during the surveying phase. The Auditors have observed that all of the larger setts missed were in difficult terrain (i.e. scrub or woodland). As the trial has progressed greater emphasis has been placed on the surveying of particularly difficult terrain by teams using radio communications. A major improvement has also been achieved by the use of a satellite based system used for the accurate identification of field locations. MAFF also recognises the benefits that accrue from a less fragmented approach to surveying by grouping farms together rather than operating within individual farm boundaries. MAFF has, where possible, already moved to this system using the same survey teams to check both sides of a farm boundary.
Sett classification
6.2 We recognise that some badger setts were mis-classified during the surveying phase and accept the Auditors' recommendation that an iterative approach be taken to sett classification and this is now current practice.
6.3 However, on the basis of ISG advice, MAFF will not take up the Auditors' recommendation that a more complex sett classification be adopted. As the ISG has observed, its value for analysis of trial data is unclear and its adoption may lead to operational confusion.
Bait marking
6.4 The Auditors recommended the use of bait marking in cases where significant doubt exists over the status of a larger sett. The use of bait-marking to delineate reactive treatment areas is being reviewed. However, it should be noted that bait marking is resource intensive and, given the seasonal variation in its value, its use extensively throughout the trial has not been feasible.
Tessellation of territories
6.5 Dirichlet tessellations are used as an aid to delineating social group territories and are not, in themselves, a means of defining boundaries. They describe areas each containing one main sett so that every point within the area is nearer to its own main sett than to any other. The tessellations can be done either manually or automatically by computer. MAFF accepts that tessellation should be a staged process and that it should be used in conjunction with limited resurveying to identify and correct anomalies in group structures. This is now established practice.
Training
6.6 MAFF accepts and has already acted in respect of the Auditors' recommendation for further training on sett classification and territory delineation procedures.
Further auditing
6.7 MAFF agrees the recommendation that the auditing of the trial should continue and has made arrangements to audit procedures in a further three triplets.
Ministry of Agriculture Fisheries and Food
February 2001
Page last modified:
12 August 2003
Page published: 5 February 2003