Promising ways to improve the lives of urban rats in South Africa

by ajmfisher16 min read20th Sep 20216 comments


Wild animal welfare

(By Andrew Fisher and Gabriele James)


This post reports on research exploring promising interventions to improve the lives of urban wild vertebrate animals in South African cities. We are grateful for the support of a grant from EA Animal Welfare Fund. The aim of this post is to present an initial catalogue of promising interventions to improve the welfare of rats in large cities in South Africa.

Rats are numerous in large South African cities (and in most cities globally), yet a relatively neglected class of urban wild vertebrate animals. The widespread use of second-generation anticoagulant rodenticides ("ARs") to control them causes them significant suffering (as well as indirect harms to non-target species). Efforts to replace ARs with more humane alternatives are therefore likely to improve rats’ welfare. 

Our research indicates that in the design, assessment, and implementation of rat management interventions, relevant regulatory bodies will need to be guided by systemic models of humaneness, be consistent with ethical principles of beneficence, and be informed by the preferences of local communities. 

In the short term, the most efficient intervention might be to ensure that rodenticides are only being used where they are genuinely required, and that these are the most humane rodenticides that are commercially available. Where possible, ARs should be replaced with more humane available alternatives. In the South African context, we believe the most viable alternatives include (i) improved rodent-proofing and environmental deterrent measures; (ii) well-designed snap traps, or electrocution traps; and failing these options, (iii) live trapping and subsequent humane despatch. While the efficacy of deterrence and rodent-proofing is well-documented, further research is required to determine the long-term effects and community buy-in for rat trapping as an alternative to ARs. 

In the medium-to-long-term, more humane lethal methods of rat population management could be evaluated in local contexts, including carbon monoxide capsules and cyanide baits. Finally, the emergence of effective and commercially available non-lethal interventions (apart from deterrence) should be evaluated. Perhaps the most promising of these emerging interventions is the use of fertility control compounds, such as SenesTech’s ContraPest. These are increasingly being adopted in the USA (but are not currently commercially available in South Africa).  

In the South African context, effective and more humane rat population management will likely require buy-in from local communities. Due to prevalent negative cultural attitudes towards rats (in some informal settlements with large-scale rat populations), this may require unconventional interventions, such as humane education. In addition, human-rat interactions appear to be more common in informal settlements, characterised by high population density; lower housing quality; and poor sanitation and waste management. These features generate a hospitable environment for rats, and may need to be addressed in order for any rat population management strategy to be sustainably effective. 


Rats are small-bodied sentient mammals which are numerous in most large cities. Rat populations (and associated human-rat interactions) are likely to be higher in low-income informal settlements of South African cities, where structural issues such as poor waste management; over-crowding; and disadvantaged socio-economic circumstances create hospitable environments for rodent populations (Roomaney et al, 2012; Jassat et al, 2012). 

The welfare of urban rats in South Africa appears to be neglected, at least relative to larger more charismatic urban wild animals such as baboons in Cape Town. In many countries rats are considered pests whose population ought to be limited as far as possible, due to the significant damage they can cause to property and food (Mason & Littin, 2003), as well as their potential threat to human health as zoonotic disease vectors (Roomaney et al, 2012). 

A representative survey of a large low-income informal settlement in Cape Town (viz Site C, Khayelitsha) suggested that residents were more concerned about the welfare impacts of second-generation anticoagulant rodenticides (ARs) on non-target species such as domestic cats and raptors, than their welfare impact on rats (Nattrass et al, 2019). A minority of surveyed residents preferred rodent control to be humane, but most did not care how rats are killed, and almost a fifth said they would be ‘happy’ if the rats suffered (indicating a worrying “pro-cruel” stance) (Buckland & Nattrass, 2019). 

Anticoagulant rodenticides (such as brodifacoum) are widely used (in block baiting stations) to control rat populations in South African cities. This is an easy and cost-effective intervention from the human perspective, but raises significant welfare concerns for rats as they typically take several days to kill, causing distress, disability and/or pain (Mason & Littin, 2003). Non-target urban wildlife species may also suffer secondary poisoning (Mason & Littin, 2003). In the local context, Serieys et al (2019) found traces of ARs in six out of seven ecologically significant predatory species in Cape Town (a recognised urban biodiversity hotspot).

Promising interventions 

General principles

  • Putative interventions should ideally be assessed and compared against a systematic model of humaneness (see e.g. Sharp & Saunders, 2011; Littin et al 2014).
  • Design and execution of interventions should be consistent with ethical principles maximising humaneness (e.g. Littin et al 2004).
  • Choice of interventions should be informed by residents’ preferences (see Khayelitsha rodent study below), to optimise uptake and associated benefits for rats’ welfare.

Use the most humane rodenticides available, and only “as necessary”

The “lowest hanging fruit” for improving rats’ welfare might be to replace currently used ARs with refined compounds that are more humane, to the extent possible. Low-cost audits could be conducted to ensure that cities are using the most humane commercially available versions of ARs (i.e. those causing the quickest and least painful deaths, and minimising the risk of suffering due to non-lethal doses). The worst effects of currently used ARs could be offset by including in the bait analgesics, anti-emetics or compounds causing sedation or unconsciousness (Mason & Littin, 2003).

ARs could also be reduced by using rodent bait stations primarily for monitoring of rat populations, and then using AR baits only when absolutely necessary to control “infestations” based on clear evidence (from monitoring) that this is required in a local area (Nattrass et al, 2019b). 

Replace ARs with more humane alternatives

Given the widespread use of relatively inhumane ARs, interventions which aim to replace ARs with more humane methods seem promising. Mason and Littin (2003) examined available rodent control interventions (in the UK), assessing them primarily in terms of humaneness for target rodents, but also considering welfare implications for non-target species. They suggest the following five forms of rodent control (as relatively humane alternatives to ARs): 

  • rodent-proofing and deterrence (as an essential long-term component of any rat control program); 
  • well-designed snap traps; 
  • electrocution traps; 
  • cyanide gas (applied to rat burrow systems); and
  • bait poison alpha-chloralose. 

Of these, alpha-chloralose and cyanide gas are unlikely to work in South African cities. Effective application of alpha-chloralose relies on temperatures below 16 degrees Celsius to induce hypothermia in rats, compromising its reliable effectiveness in more temperate South African cities. Cyanide gas (and other less humane fumigants) have welfare advantages (viz dependent young are not left to die in the nest because all animals in the burrow are killed at the same time, and low risk of secondary poisoning to non-target animals). However, fumigants are typically applied to control outdoor rats in burrow systems, so their application is not indicated for the most severe “infestations” of indoor rats prevalent in informal settlements of South African cities. Further, fumigation is usually expensive and requires regulation (i.e. the use of special equipment by a licensed operator); burrow systems need to be found which is labour-intensive; for human safety reasons the method cannot be used in domestic settings where human-rat interactions are highest (especially if not well-regulated or monitored), nor where the soil is sandy or loose (e.g. in the Cape Flats where the large informal settlement of Khayelitsha is situated). 

For these reasons, we think the most promising alternatives to ARs in South African cities are likely to be:

  • improved rodent-proofing and deterrence (as an essential component); alongside:
  • well-designed snap traps; or electrocution traps;
  • failing these, live trapping and subsequent humane despatch (rather than releasing rats away from human settlements, which has worse welfare impacts (Mason & Littin, 2003)).

Adequate rodent-proofing and deterrence are essential long-term sustainable interventions, given that more humane alternatives to ARs only work if carried out continuously, due to the swiftness with which rat populations recover (Mason & Littin, 2003). This is especially important in the South African context, where high poverty, inequality and unemployment drive expansion of urban informal settlements characterised by erratic and inadequate waste disposal systems that attract rats (Green, 2019). General strategies include rodent-proofing housing and waste disposal systems; limiting access to harbourage, food and water; and introducing predators (both wild and domesticated) (Mason & Littin, 2003). More concrete interventions might include reducing food waste; improving refuse collection and disposal systems; improving quality of housing (e.g. blocking holes where rats can live); and improving sanitation systems. Whilst improving housing and sanitation; and eliminating access to food, water and harbourage, seem uncontroversial from a rat welfare perspective, we are hesitant to recommend the introduction of predators unless and until the overall welfare implications for rats (and non-target species) have been thoroughly investigated.

Where traps are used to replace ARs, well-designed and high-quality snap traps should be used to minimise risk of non-lethal injury to (target and non-target) animals. There is wide variation in the mechanical performance of snap traps (used as a proxy for humaneness), so any use of snap traps as part of rodent control programs in South African cities should preferably be subject to regulatory approval (e.g. in line with the UK trap approval process) (Baker, 2012). Electrocution traps could also be used, but only those that cause instant stunning. Failing the first line use of snap or electrocution traps, live trapping might be acceptably humane if traps are well-monitored to minimise the time spent trapped, and the despatch (rather than release) of trapped rats is as rapid and humane as possible (Mason & Littin, 2003).

The replacement of ARs with traps can be labour-intensive because large numbers of traps are usually needed, and they require regular checking (Mason & Littin, 2003); however this could be an advantage in low-income informal settlements where trap-setting and monitoring has been a source of employment for low-skilled workers (see Khayelitsha rodent study, below). Traps also have welfare disadvantages e.g. adult females but not their nestlings are killed, leaving pups to die of starvation; and non-target animals could also become trapped (though this risk can be mitigated if traps are well-designed and appropriately placed) (Mason & Littin, 2003).

A survey of the use of snap traps in Khayelitsha suggested residents’ concerns that traps might injure children and household pets, and that they would have to deal with the unpleasant mess of killed rats in traps (Nattrass et al, 2018). According to local environmental health officials, residents preferred dealing with drowned rats than with rats mutilated by traps (Nattrass et al, 2018), which could partly explain why about a third of households continued to use rodenticides even when provided with snap traps (as part of a study of the acceptability of such traps in poor urban communities) (Roomaney et al 2012). Nevertheless, that acceptability study found that 88% of respondents used the traps to catch rats while decreasing their reliance on previously used illegal street rodenticides (Roomaney et al 2012). The acceptability study also highlighted that the relative expensiveness of snap traps was a barrier to their perceived acceptability, so perhaps government subsidies or incentives would be needed to overcome this obstacle to adoption. 

In any event, the replacement of ARs with traps would need to be complemented with addressing the causes of “infestations” in informal settlements e.g. improving sanitation and waste control (see “Rodent-proofing and deterrence”, above). More research is needed to determine if sustained snap trapping in poor urban communities reduces rat populations and use of ARs (and other illegal street rodenticides) in the long-term (Roomaney et al, 2012).

Investigate the effectiveness of new (more humane) lethal methods

In the medium-to-long term, potentially more humane new lethal methods could be evaluated. These might include carbon monoxide capsules, or cyanide baits, both of which might be more humane, carry a lower risk of sub-lethal dosing (compared to fumigants), and lower the risk of poisoning to non-target species or humans (compared to fumigants) (Mason & Littin, 2003). If found to be more humane and comparably effective, they could then be developed and used as replacements for currently used ARs.  

Replace ARs with non-lethal methods (besides deterrence)

Non-lethal fertility control compounds look promising as an alternative to lethal methods (Witmer & Raymond-Whish, 2021; Siers et al 2017); however more research is needed to evaluate its effectiveness and humaneness. SenesTech’s ContraPest is widely used in some areas (e.g. California) and appears to be effective in reducing the fecundity of exposed rodents under certain conditions, with potentially acceptable humaneness (Witmer & Raymond-Whish, 2021). Although (as far as we are aware) fertility control compounds like ContraPest are not yet commercially available in South Africa, there is some indication that environmental health officials in Cape Town would be open to fertility control as an alternative to ARs (Nattrass et al, 2019). However, although promising in principle, replacement of ARs with a fertility control program is unlikely to be considered viable by South African cities in the short-to-medium term (even if it became commercially available). This is plausibly due to its relative expensiveness and budgetary constraints, in a context where provision of basic services and addressing poverty, unemployment and inequality are pressing.

Other potential non-lethal alternatives to ARs (and other more humane lethal methods) include developing non-lethal repellent compounds e.g. (synthetic analogues of) predator odours aversive to rodents, with low rates of habituation (Mason & Littin, 2003). 

Local considerations from the Khayelitsha rodent study

Nattrass et al (2019) studied a pilot program by Cape Town Environmental Health (EH) which employed previously un-employed residents of a large informal settlement in Cape Town (Site C, Khayelitsha) to live trap and drown rats (“Khayelitsha rodent study”). This was intended as an alternative to the use of ARs, which were seen by EH officials as posing a risk of secondary poisoning to children and (domestic and wild) non-target animals. Controversy arose when the National SPCA overturned the program on the ground that drowning rats was inhumane and illegal, forcing the City to return to using ARs. Despite this, most residents thought that the cage-trapping and drowning project should go ahead. This research also suggests that live-trapping and drowning of rats in informal settlements might be more humane than the use of ARs, although this is unclear due to the distress of trapped rats prior to drowning (more research is needed to clarify this). The overall welfare impacts on rats (of replacing ARs with “live trap and drown” programs) needs to be further investigated before we recommend re-implementation. 

The Khayelitsha rodent study also yielded helpful insights for selecting interventions in the context of South African cities characterised by rat-attracting informal settlements. The study confirmed that rat populations are greater in informal settlements than in formal housing areas, due to greater housing density, lower housing quality and more erratic rubbish removal in informal areas. A key driver of high rat populations in these areas is failures in the refuse collection system: over a third of households in the shack areas reported that their main method of disposal of rubbish was to dump it on the street or in the river (Green, 2019). Further, people did not always receive their allocated two rubbish bags per week, dogs often opened up rubbish bags before they were collected, and rubbish containers were often used for other purposes (e.g. car wash, drug dealing) (Green, 2019). This suggests that reforming the current waste management system in large informal settlements, as well as expanding education campaigns about waste control and rat-proofing, could produce net gains for rat welfare in informal settlements (Green, 2019). 

The study also identified factors of residents’ attitudes towards cruel or humane rodent control. Most people said they did not care how rats were killed, although a minority said they would prefer rats not to suffer (pro-humane stance) and approx. 19% said they would be happy if the rats did suffer (pro-cruel stance) (Buckland & Nattrass, 2019). Those concerned that rats might be linked to witchcraft were more likely to have a pro-cruel stance, while those who agreed that animals should be treated kindly were more likely to have a pro-humane stance (Buckland & Nattrass, 2019). These findings suggest that introducing more humane alternatives to ARs in informal settlements would be influenced by context-specific factors shaping residents’ attitudes towards rat welfare such as cultural beliefs about witchcraft. For example, interventions might need to be accompanied by school-level education campaigns aimed at increasing rates of pro-humane attitudes towards rats (Buckland & Nattrass, 2019). Effective interventions might require complementing improved sanitation and housing with efforts to influence public attitudes to increase support for animal welfare and reduce culture-linked fears about rats being associated with witchcraft (Buckland & Nattrass, 2019). 


Baker, S. E., Ellwood, S. A., Tagarielli, V. L., & Macdonald, D. W. (2012). Mechanical performance of rat, mouse and mole spring traps, and possible implications for welfare performance. PloS one7(6), e39334.

Buckland, A., & Nattrass, N. (2020). Understanding Preferences for Humane and Cruel Treatment of Pest Rodents in Site C, Khayelitsha, South Africa. Journal of applied animal welfare science23(3), 315-324. 

Green, E. (2019).  Why is rubbish building up in Khayelitsha? An investigation into the contestation over Khayelitsha’s refuse removal service (CSSR Working Paper No. 427). Centre for Social Science Research, UCT. 

Jassat, W., Naicker, N., Naidoo, S., & Mathee, A. (2013). Rodent control in urban communities in Johannesburg, South Africa: From research to action. International Journal of Environmental Health Research23(6), 474-483.

Littin, K., Fisher, P., Beausoleil, N. J., & Sharp, T. (2014). Welfare aspects of vertebrate pest control and culling: ranking control techniques for humaneness. Revue scientifique et technique (International Office of Epizootics)33(1), 281-9.

Littin, K. E., Mellor, D. J., Warburton, B., & Eason, C. T. (2004). Animal welfare and ethical issues relevant to the humane control of vertebrate pests. New Zealand Veterinary Journal52(1), 1-10.

Mason, G. J., & Littin, K. E. (2003). The humaneness of rodent pest control. Animal Welfare 12(1). 

Nattrass, N., Stephens, J., & Loubser, J. J. (2019). Animal welfare and ecology in the contested ethics of rodent control in Cape Town. Journal of Urban Ecology5(1). 

Nattrass, N., Stephens, J., & Loubser, J. J. (2018). The Rat Trap: Contestation over rodent control in Cape Town. (CSSR Working Paper No. 410). Centre for Social Science Research, UCT. 

Nattrass, N., O’Riain, J., Bishop, J., Rogan, M., Serieys, L. (2019b). How rubbish and rat poison threatens wildlife at the University of Cape Town, and what we can do about it. iCWild working paper

Roomaney, R., Ehrlich, R., & Rother, H. A. (2012). The acceptability of rat trap use over pesticides for rodent control in two poor urban communities in South Africa. Environmental Health11(1), 1-6. 

Serieys, L. E., Bishop, J., Okes, N., Broadfield, J., Winterton, D. J., Poppenga, R. H., ... & O'Riain, M. J. (2019). Widespread anticoagulant poison exposure in predators in a rapidly growing South African city. Science of the Total Environment666, 581-590.

Sharp, T., & Saunders, G. (2011). A model for assessing the relative humaneness of pest animal control methods. Canberra, Australia: Department of Agriculture, Fisheries and Forestry.

Siers, S., B. Pyzyna, L. Mayer, C. Dyer, I. Leinbach, R. Sugihara, and G. Witmer (2017). Laboratory evaluation of the effectiveness of the fertility control bait ContraPest on wild-captured back rats (Rattus rattus) (Final Report QA-2570). U.S. Department of Agriculture, Animal and Plant Health Inspection Service, National Wildlife Research Center, Hilo, Hawaii, USA.

Witmer, G. W., & Raymond-Whish, S. (2021). Reduced fecundity in free-ranging Norway rats after baiting with a liquid fertility control bait. Human–Wildlife Interactions15(1), 17.


6 comments, sorted by Highlighting new comments since Today at 7:08 AM
New Comment

Fantastic post and research!

I don't know about rats much but had them as pets in elementary school (they are quite affectionate) and they shouldn’t suffer like this.

This may not be in scope of your article, but here’s some questions that might be valuable to answer:

1. What is the total population of urban rats at stake in your area of South Africa? E.g. X number of animals and how many die poorly? 

2. In your area, what is the total size of vertebrate urban wildlife and what is the relative number of rats (compared to pigeons, mice, etc.)? 

3. Why choose rats? Any particular reason?

4. Any guesses on the cost effectiveness of interventions? (maybe especially compared to established EA animal welfare interventions?) Cost effectiveness is difficult and important, so even a sketch would be useful, e.g. saying something like "Private conversations with experts suggested there is a sense that policy campaigns can be low cost" is useful.

5. Is there any "flow-through"/learning/"capacity building" value from executing the mentioned interventions on other animal welfare projects, especially worldwide, outside of South Africa?


Maybe you know the answer to some of the above, or maybe you have a link to someone who does. Or if you don't think anyone has an answer right now, it’s good to know this too!

I think answers to questions 4 and 5 would be very interesting and helpful to a new org or someone in this space.

Great article, this seems like a really important area!

Hi Charles, 

Thanks for taking the time to read our article. You raise some good questions.

  1. Concrete data on urban rat populations is very scarce, partly due to the difficulty and labour intensiveness of gathering data. Even when these studies are conducted, the numbers tend to be broad estimates based on nuisance reports or capture-recapture methods. For example, in 2014 the estimated rat population of New York City was around 2 million . Considering that Cape Town has a smaller population and lower population density than NYC, it is plausible that it has a smaller rat population. However, based on the frequent news coverage of rat infestations, especially in informal settlements, and the number of bait stations maintained by the city (over 10, 000 in 2020) it is likely that the population is significant. 
  2. This is a great question, but again one that is difficult to answer, for the reasons outlined above, when it comes to populous small vertebrates (especially if they are considered pests or alien species and therefore not an ecological priority). The population s of more charismatic urban wildlife species are better studied. For e.g. there are an estimated caracals 45,000 to 150,000 across the entire country, and about 500 baboons in the Cape.  
  3. Rats are numerous, their welfare is neglected, and the negative impact on their welfare due to the prevailing forms of population management are significant.
  4. I don’t know the full answer to this one, but here is what I’ve come up with so far. I think many of the alternatives proposed here don’t compare very favourably from a cost-efficacy perspective to ARs. Alpha-chloralose, for example, is more expensive than ARs (Mason and Littin), and rodent-proofing can also be costly and time consuming. However, there is some evidence that snap traps can be favourably cost-effective compared to ARs (Mason and Littin). The Khayelitsha program (using live trapping) was also comparably low-cost to ARs . With regard to the development of more humane lethal methods, I am guessing that the research and development costs would be quiet significant.  
  5. Our research so far suggests that some other cities in South Africa and in Africa use a similar rat management strategy to Cape Town, and in some instances have similar environmental and socio-economic drivers of their rat populations. In these instances the interventions and considerations explored in this research may be a valuable starting point. 

Interesting! Potentially ignorant question, from someone who loves rats and humans: Do urban rats spread disease and/or bite people? I was surprised that those potential risks didn't figure into the survey or people’s responses and attitudes. What are the risks to the humans of letting rats live among them? If there are real health risks to people from having a large rat population around, I would expect a culture would be wise to use stories of witchcraft to spread the wisdom that rats are “bad news” to human health. It’s sad when one mammal’s welfare (humans) is at odds with another mammal’s welfare (rats) but I feel more trust in a species-specific welfare agenda when I have info about the trade offs to other species in their ecosystem.

I think your comment was intended for the original posters, Gabriele James or Andrew Fisher, either of whom know way more than me, since my knowledge of this is from reading this post.

You might want to directly reply to their post so they can see your thoughts.

Uh, some quick thoughts anyways:

It’s sad when one mammal’s welfare (humans) is at odds with another mammal’s welfare (rats) but I feel more trust in a species-specific welfare agenda when I have info about the trade offs to other species in their ecosystem.

I think your premise is that there has to be a trade off between rats or humans, but I don’t think this is the premise of likely interventions or the view of the original posters.

Importantly, I think one intervention is the reasonable reduction of urban rat populations, e.g. removing garbage and food sources, so we don’t have a lot of rats who suffer.

(Note that there’s a little micro-history in wild animal welfare where some people got the sense that the outcome of the whole idea was removing animal populations in general, but this is basically too simplistic. As far as I can tell, wild animal welfare people like animals and nature, you know, like normal people do).

Do urban rats spread disease and/or bite people? I was surprised that those potential risks didn't figure into the survey or people’s responses and attitudes. What are the risks to the humans of letting rats live among them? If there are real health risks to people from having a large rat population around, I would expect a culture would be wise to use stories of witchcraft to spread the wisdom that rats are “bad news” to human health.

Well, South Africa has a really vibrant culture and I can’t imagine how to characterize it, but well, my guess is that there is strong sentiment against rats among some.

As the OP said, there was “pro-suffering”, people wanted to see rats suffer (despite the fact rats have no control and literally just want to eat to survive).

This is really cool! Love seeing tangible efforts for near-term and large-scale reduction in animal suffering. The lack of easy and cost-comparative alternatives to ARs worry me a bit, though.

I suspect the "ARs causes secondary poisoning to pets" angle would more effectively convince governments to mandate using alternatives. I don't know if they have much power to enforce it, though.

General strategies include rodent-proofing housing and waste disposal systems; limiting access to harbourage, food and water; and introducing predators (both wild and domesticated)

Do you guys think introducing predators (presumably cats) is a welfare improvement over ARs?

Hi Emrik, 

Thanks so much for reading the article. I'm glad you found it interesting! I'm also concerned about the costliness of more humane alternatives, but perhaps there is some hope that some of the costs could be driven down if there was enough adoption to scale production. Some of the interventions may also be more costly in the short-term but prove less expensive than ARs in the long-term. In this case I have in mind things like rodent-proofing and deterrence. 

I agree with you too that from an advocacy perspective, alternatives to ARs are more likely to gain public support by appealing to the welfare costs of ARs to non-target beings like pets, children, and wild animals that aren't considered pests. 

I don't have a definite answer about domestic cats, unfortunately. As a cat companion, I am often disturbed by the  way they slowly hunt and kill their prey, but this may still be better than dying from ARs. On a more serious note, from what I've read and heard, one of the complications regarding domestic cats is their impact on non-target wildlife. Some of these costs could be mitigated using bell-collars and keeping cats indoors, although especially in informal settlements this would not be easy to regulate.