Guidelines on implementing the 3Rs in both field and captive studies.
On this page
A wide variety of wildlife is used in research, (including mammals, birds, reptiles, fishes, amphibians and invertebrates) in studies aimed at:
- Understanding species behaviour and ecology
- Species conservation
- Population management
- Evaluating methodologies for control
- Understanding the role of wildlife in disease transmission.
Wildlife studies vary in their invasiveness and impact on the animals being studied . In all circumstances, researchers should seek to minimise any negative impact on the welfare of animals involved. Good animal welfare practice for wildlife research is characterised by the same features as laboratory-based research, however different approaches and procedures may be needed for wild animals compared with laboratory-bred animals.
Wildlife research is usually conducted with free-living animals in their natural habitat or with wild-caught animals in various captive settings (e.g. laboratory, zoo, aquarium, sanctuary). Rarely, wildlife species are purpose-bred under laboratory conditions similar to those used for animal models. Animals should not be taken from their natural habitat unless animals bred in captivity are unavailable or unsuitable for the scientific purpose. Taking animals from the wild for scientific purposes is regulated by legislation.
Many wildlife studies focus on conservation and management, with the aim of learning about the ecology of a population in the field. In such cases, minimising disturbance to the animals is important for the scientific validity of the research as well as for good animal welfare [2-4].
Some field studies require altering the animals' habitat or behaviour as a goal of the study, whilst others require monitoring the animals in response to a change in habitat. In such cases, it is important to minimise disturbance, both to the animals around the study site and to the animals under investigation.
Many field studies involve manipulating the study animals involving capture, marking or additional procedures, or a combination of these, which can cause distress. Capture, marking, radio tagging and collecting physiological data (e.g. blood or tissue samples) can also have delayed consequences, such as a reduced probability of survival and reproduction. It is therefore vitally important to carry out such procedures according to 'best practice' and to monitor the animals for potential adverse effects. Pilot studies may be used to assess the potential environmental disruption of fieldwork, which should be minimised, and follow-up studies may be used to monitor the success of the study and any adverse effects caused to the animals.
Researchers should take into account the social structure and behaviour of the species under investigation. The most obvious example is the dependence of young on maternal care. For species with a complex social organisation, removing a critical member of the social group can impair the well-being of the remaining group members. Such considerations may be pertinent even when the removal of animals is temporary.
Even purely observational studies, where there is no manipulation of the animals, can raise ethical concerns with regard to animal welfare and/or conservation. For example, human observation can disturb normal animal activities such that animals abandon their territories, home ranges or young. Making trails/transects through habitats to access, observe and census animals can also cause disturbance. Researchers should consider such issues when designing their studies. Camera traps can sometimes be used to avoid disturbing the animals either by trapping or direct observation.
An animal's behaviour is influenced by its genetic make-up, experience and social background, and these can be potential sources of sampling bias in both field and captive studies with wildlife species. The STRANGE framework provides a means for researchers to avoid sampling bias by considering these issues during the design, conduct, reporting and interpretation of behavioural studies.
In general, wild-caught animals should be kept in captive conditions that conform as closely as possible to their natural habitat (e.g. in such respects as light intensity, food etc.). Length of time in captivity and location of release are important additional considerations for animals being returned to the wild. Prolonged time in captivity may result in the released animal being rejected by its conspecifics and losing access to essential resources (e.g. badger's sett), which may compromise its ability to feed or fend for itself. Animals should always be returned to the exact point of capture.
The 3Rs are an integral part of the UK Animals (Scientific Procedures) Act 1986, amended 2012 (ASPA), which regulates the use of vertebrates and cephalopods, including wildlife species, in procedures with the potential to cause pain, suffering, distress or lasting harm. The ASPA also requires that the likely benefits of the research, to humans, animals or the environment, are weighed against the likely harms to the animals involved.
The 3Rs and harm/benefit assessment are relevant also to wildlife research that is not regulated by the ASPA but which, nonetheless, has the potential to compromise the welfare of the study or non-study species. The 3Rs should always be considered as part of the design and conduct of wildlife studies [2,3,5].
Replacement does not often apply to studies aimed at understanding the behaviour and ecology of wildlife species, because the animals themselves are the objects of study. However, in silico techniques, such as computer modelling, are used for population studies, including those aimed at evaluating methods of lethal/fertility control, investigating animal movements and predicting disease spread.
Many of the principles and techniques used to reduce the numbers of animals used in biomedical research are applicable to wildlife research. These include:
- Using appropriate and efficient experimental designs, e.g. factorial designs to explore the effects of several variables in one experiment; sequential and multivariate statistical methods; repeated measures designs; phasing of experiments.
- Estimating the minimum sample size necessary for adequate statistical power using information from the literature, pilot data, or both.
- Keeping the number of replicates/experimental units (e.g. individual animals, cages of animals, social groups) to the minimum for the power level required.
- Utilising statistical programmes that indicate when sufficient data have been collected.
- Avoiding repeating studies, unless it is essential for the purpose or design of the project.
- Sharing data and resources (e.g. biological and genetic samples) and publishing results, preferably in free access formats.
Animal use can also be reduced by:
- Minimising the number of procedures to be carried out (e.g. trapping once to gain data for different parts of the study).
- Utilising species/sex/age-specific experimental designs (e.g. using species specific baits, or trapping at specific times or locations to minimise non-target capture).
In contrast to most laboratory studies, sample size is not easy to control in field studies. For example, it may be necessary to trap 100 animals to find 40 that meet the age and sex requirements for a study. In addition, there may be external factors, such as weather conditions, that may affect the data that can be collected. It is important to consider these factors when designing field studies.
Anyone capturing animals should be trained and competent in humane methods of capture, handling and release, and in any scientific procedures used, to minimise the impact on animals and their environment.
Wildlife species used in research vary greatly in their body size, physiology and behaviour. The responses of animals to disturbances, handling and procedures can vary, both within species (e.g seasonally) and between species. Professional guidance, species-specific expertise and species appropriate literature should therefore be consulted to ensure the methods and equipment are appropriate and that up-to-date refinements and best practice are implemented.
Assessing potential sources of harm to study and non-study species and how these will be eliminated or minimised should form part of all wildlife research proposals. In many cases, any negative impact on animal welfare can be reduced by careful planning, robust experimental design and choosing the least invasive techniques for the species in question. Issues to consider include:
Capture/trapping – best practice guidelines are available (see professional guidance below); smart methods of monitoring the location of tagged animals (e.g. spatial loggers) avoid the need for repeated trapping; use of intelligent traps will prevent unnecessary recaptures and the trapping of vulnerable non-target animals. Maintain an awareness of local weather; extreme conditions can prevent traps from being checked in a timely manner and flooding or high temperatures place trapped animals at risk.
Handling and restraint – handling wild animals should always be kept to a minimum; the correct techniques are not necessarily the same as those recommended for laboratory animals.
Marking methods (e.g. banding; tagging; tattooing; toe, ear or tail clipping) – the least painful or distressing methods should be used (see table below).
Attaching/implanting tracking devices such as radio transmitter – the device should be within the recommended percentage of weight of the study animal and also be suitable for the animal's lifestyle (e.g. external transmitters may not be suitable for animals that squeeze through small openings, such as bats). In addition to animal welfare, any effects on the behaviours under study should also be considered (e.g. birds wearing tracking devices have increased energy expenditure and are less likely to nest ).
Medical/surgical interventions using anaesthetics and immobilising agents – anaesthesia is often needed to prevent distress or injury when handling or fitting non-invasive devices (e.g. radio collars).
Sample collection (e.g. hair, feathers, scales, milk, skin scrapings, stomach contents) – even non-invasive methods can be refined to reduce stress/distress (see table below).
Sampling of blood – if alternatives, such as saliva or faeces, cannot be used in place of blood or plasma, what can be done to minimise the impact to the animal? Will fluids be required to minimise the risk of dehydration? Can microsampling be used to reduce blood loss?
Methods for measuring body weight, respiration rate, heart rate, pulse rate, body temperature and body lengths – what steps can be taken to minimise stress during these procedures?
Euthanasia – humane killing methods in the field are not necessarily the same as those in Schedule 1 of the ASPA; researchers should be trained in the most appropriate field methods for the animals likely to be encountered (i.e. not just the target species).
Close examination of den sites, nests, etc. – could this trigger abandonment, infanticide, increase predation risk, or otherwise impact on offspring survival? How can disturbance to the animals be minimised?
Transport – guidance on minimising stress is given on our transport page.
Housing and maintenance in captivity – for most wild animal species there are no specific guidelines in the Home Office code of practice on animal care and accommodation; specialist advice should be sought on captive maintenance; consider outdoor housing where appropriate.
Environmental and social manipulation – longer term impacts on individuals and communities should be carefully considered; disturbances during key parts of the life cycle should be minimised.
Social deprivation – group-living species can suffer considerable stress if held in isolation.
Changes to diet and access to food/water – wild animals may not be able to use drinking bottles or to recognise certain novel foods; knowledge of the natural diet and behaviour and close monitoring of food and water intake is essential.
Methods of attracting animals (e.g. playback of calls, provisioning, baiting) – care should be taken to tailor the method to the target animals and to minimise wider or longer-term impacts on communities, such as disruption to breeding or reliance on provisions.
Frequency and duration of human observation – consider using remote video surveillance.
Habituation to humans – this is a particularly important factor to consider where humans may pose a threat to wildlife (e.g. through hunting).
Disturbing interactions between species (e.g predator-prey), within species (e.g competition), and between species and habitat – such interactions can be complex and the potential knock-on effects of disturbances should always be considered.
Required data: traditional method
Physiology: blood sampling
Diet: lethal sampling (stomach contents)
ID: invasive marking, attachment of instruments
Less invasive marking (e.g. dyes)
Genetics: tissue sampling
Swabbing (e.g. the mucus layer of fishes and cephalopods; bats’ wings)
Discarded egg shells or feathers
Animals (Scientific Procedures) Act
The ASPA defines a regulated procedure as anything done to a protected animal (vertebrates and Octopus vulgaris) for a scientific purpose that causes pain, suffering, distress and/or lasting harm. This includes any material disturbance to normal physical, mental and social well-being, as well as disease, injury and physiological or psychological discomfort either immediately or in the long-term. It is essential to contact the Home Office Inspectorate where there is any doubt whether a procedure falls within the scope of the ASPA.
Ringing, tagging or marking an animal primarily to identify it as a specific individual, or using any other humane way to do so, are not regulated procedures if they cause no more than momentary pain and no lasting harm. For example, micro-chipping or ear-marking a rodent is not a regulated procedure if it is being done primarily to identify the animal. Similarly, blood or DNA sampling solely to establish the identity or provenance of an animal would not be regulated if the intervention caused no more than momentary discomfort or distress. Methods of marking or identification, such as toe clipping, which can cause suffering in excess of this threshold, are regulated when carried out for an experimental or other scientific purpose. Always check with the Home Office Inspectorate if intending to carry out any of the techniques listed above. It may be that, for welfare reasons, anaesthetics should be used in order to minimise distress and injury, which would require regulation under ASPA. See the Guidance on the Operation of the ASPA for more information.
National laws, e.g. the UK Wildlife and Countryside Act 1981, regulate the taking of animals from the wild for scientific purposes. All wild mammals are also protected from certain acts of cruelty under the Wild Mammals (Protection) Act 1996. Badgers, deer and seals have Acts of Parliament dedicated especially to them. Investigators should ensure that they comply with all relevant regulations and that they possess all the necessary permits and licences. Note different bodies are responsible for issuing of licensing and provision of wildlife management advice in the devolved regions of the UK.
Many wildlife species are threatened or endangered and protected by the Convention on International Trade in Endangered Species of Wild Flora and Fauna (CITES). The CITES database and World Conservation Union (IUCN) Red List of Threatened Species can be used to ascertain the current conservation status of a given species. CITES Appendices stipulate that national Government permits are required for trade in certain species and provide three levels of protection for those species in international commercial trade.
Endangered or threatened species should not be used in research unless there is special justification, e.g. a direct conservation or welfare application. The ASPA places restrictions on the purposes for which endangered animals (including non-human primates) can be used.
Guidance on the use of wild species is also available from various professional organisations and societies, including the American Society of Mammalogists, American Society of Ichthyologists and Herpetologists, The Ornithological Council, Association for the Study of Animal Behaviour, Canadian Council on Animal Care and The Chartered Institute of Ecology and Environmental Management (CIEEM).
Reporting wildlife research
Transparent reporting of how a study was designed and carried out allows others to replicate or adapt the methods and assess the reliability of the findings. The need for increased transparency in wildlife research reporting has been highlighted for camera trap [10,11] and drone  use. Thorough and open reporting in wildlife research also furthers our understanding of how variations in study design can impact upon the behaviour and ecology of species, such as birds fitted with tracking devices [6,13]. This information can be used to refine future studies, contributing to improvements in animal welfare and data quality.
Good reporting is facilitated by effective planning, and both contribute to rigorous and reliable research. The ARRIVE and PREPARE guidelines are resources created to help researchers effectively plan (ARRIVE and PREPARE) and transparently report (ARRIVE) research involving animals. The ARRIVE and PREPARE guidelines are endorsed by the Association for the Study of Animal Behaviour (ASAB) and the Animal Behavior Society (ABS). Using the ARRIVE guidelines for reporting is also recommended by journals including Nature, Animal Behaviour, Animal Ecology in Focus and International Journal of Primatology.
Using the ARRIVE guidelines for wildlife research
The ARRIVE guidelines outline the essential information that should be included when writing a manuscript that describes animal research. ARRIVE-compliant reporting does not require certain procedures or study designs to have been followed, rather it requires sufficient information on what procedures and study designs were used.
The quality of reporting, reliability and reproducibility of wildlife research is improved by inclusion of all relevant items within the ARRIVE 2.0 Essential 10 and Recommended Set. The majority of examples provided within the ARRIVE explanation and elaboration document  have a preclinical focus, but most ARRIVE guidelines items are relevant across all fields of animal research. The below examples outline a selection of items from the ARRIVE Essential 10 in the context of wildlife research reporting.
Depending on what you plan to make inferences about, include a justification for the number of individual organisms, animal groups and/or field sites included within your study. Compliance with ARRIVE does not require a sample size calculation to be carried out, rather it requires an explanation of how the animal numbers were determined.
- Estimates based on previous census data or expected species density .
- Limitations resulting from sampling effort, adverse events or permit stipulations .
- Calculations for achieving sufficient power for statistical hypothesis testing .
Inclusion and exclusion criteria
Include details on any eligibility or disqualification criteria applied to individual organisms, animal groups, field sites or other areas of data collection. For example:
Animals with a bodyweight of <2g were not included in the study due to a higher risk of trapping-associated mortality.
Sampling occurred between September and February to prevent disruption of breeding behaviour.
Eight potential field sites were identified using our pilot data.From this eight, three were selected based on their proximity to facilities that could be used to analyse samples.
Provide details on the measures taken to minimise the influence of bias and confounds. State whether randomisation was implemented at any stage of the study, and if so how. If randomisation was not appropriate for the study, state what methods were used for selection and allocation.
To estimate animal abundance, transect placement was stratified to allow for regular tessellation of the study site (i.e. not randomly placed) following the recommendations of .
Within the study area, each pasture (the experimental unit) was randomly assigned to a treatment (intensive or organic management) using random numbers generated by the standard = RAND() function in Microsoft Excel.
To minimise confounds associated with the time of day, study sites were visited in a random order following a schedule created using a computer-based random order generator.
Blinding is another measure that can be incorporated into a study design to minimise unconscious bias. Report whether blinding was incorporated into the study design at any stage. For example:
Blinding was incorporated into the data collection stage. The primary observer was kept unaware as to which animals were wild born and which were reintroduced as part of captive breeding programmes.
It was not possible to incorporate blinding at the data collection stage due to the obvious phenotypic differences between groups. However, data were then coded to allow for blinding during data analysis.
- Zemanova MA (2019). Poor implementation of non-invasive sampling in wildlife genetics studies. Rethinking Ecology 4: 119. doi: 10.3897/rethinkingecology.4.32751.
- Soulsbury CD et al. (2020). The welfare and ethics of research involving wild animals: a primer. Methods in Ecology and Evolution, 11(10), 1164-81. doi: 10.1111/2041-210X.13435.
- Lindsjö J et al. (2016). Animal welfare from mouse to moose - implementing the prinicples of the 3Rs in wildlife research. Journal of Wildlife Diseases 52(2): 65-77. doi: 10.7589/52.2S.S65.
- Field KA et al. (2019). Publication reform to safeguard wildlife from researcher harm. PLoS biology 17(4): e3000193. doi: 10.1371/journal.pbio.3000752
- Zemanova MA (2020). Towards more compassionate wildlife research through the 3Rs principles: Moving from invasive to non-invasive methods. Wildlife Biology 1. doi: 10.2981/wlb.00607.
- Barron DG et al. (2010). Meta‐analysis of transmitter effects on avian behaviour and ecology. Methods in Ecology and Evolution 1(2): 180-187. doi: 10.1111/j.2041-210X.2010.00013.x
- Bogucki R et al. (2018). Applying deep learning to right whale photo identification. Conservation Biology 33(3):676-684. doi: 10.1111/cobi.13226
- Adams CI et al. (2019). Beyond biodiversity: Can environmental DNA (eDNA) cut it as a population genetics tool?. Genes 10(3): 192. doi: 10.3390%2Fgenes10030192
- Blackman RC et al. (2021). Mapping biodiversity hotspots of fish communities in subtropical streams through environmental DNA. Scientific reports 11(1): 1-11. doi: 10.1038/s41598-021-89942-6
- Choo YR et al. (2020). Best practices for reporting individual identification using camera trap photographs. Global Ecology and Conservation 24: e01294. doi: 10.1016/j.gecco.2020.e0129
- Burton AC et al. (2015). Wildlife camera trapping: a review and recommendations for linking surveys to ecological processes. Journal of Applied Ecology 52(3): 675-85. doi: 10.1111/1365-2664.12432
- Barnas AF et al. (2020). A standardized protocol for reporting methods when using drones for wildlife research. Journal of Unmanned Vehicle Systems 8(2): 89-98. doi: 10.1139/juvs-2019-0011
- Geen G. R et al. (2019). Effects of tracking devices on individual birds–a review of the evidence. Journal of Avian Biology: 50(2): e01823. doi: 10.1111/jav.01823
- Reynolds PS (2019). When power calculations won’t do: Fermi approximation of animal numbers. Lab Animal 48: 249–53. doi: 10.1038/s41684-019-0370-2 Percie du Sert N et al. (2020). Reporting animal research: Explanation and elaboration for the ARRIVE guidelines 2.0. PLoS biology 18.7: e3000411. doi: 10.1371/journal.pbio.3000411
- Morrison ML et al. (2008). Wildlife Study Design. 2nd edition. Springer.
- Barabesi L and Fattorini L (2013). Random versus stratified location of transects or points in distance sampling: theoretical results and practical considerations. Environmental and Ecological Statistics 20(2): 215-36. doi: 10.1007/s10651-012-0216-1