Stoat Control in Practice


Most official stoat trapping operations since 1972 have used Fenn traps (pp. 326, 327), set in tunnels and baited with hens' eggs or meat. Fenns are practical, easily operated, and well documented (Anon. 1981), and still offer the only proven and safe method of removing stoats over wide areas (Dilks et al. 1996, 2003; Lawrence & O'Donnell 1999). Vertical bars set across tunnel entrances (King & Edgar 1977) at 30 mm spacing discourage entry by ground-feeding birds while allowing stoats to pass through at full speed (Short & Reynolds 2001). Conibear traps and various other types have been tried, but are not widely used.

In recent years more attention has been given to the question of balancing the need of native populations for protection with the need of individual pest animals for humane treatment (Littin et al. 2004). New standards for kill traps have been developed, including the requirement that captured animals should be rendered irreversibly unconscious within 3 minutes. The Fenn trap does not pass this test (Purdey et al. 2004), so it will eventually be replaced by more recent trap designs.

On the other hand, the design of a single-catch trap is only one component of effective control by trapping, for two reasons. First, the average probability of capture is relatively low, often less than 0.2/day, and is zero for at least some individuals (King et al. 2003a). In areas where traps are set for years at a time, some stoats learn to become very wary of traps and resistant to capture even if dozens of traps are concentrated over a limited area (Chapter 8). Second, even if efficiency is maximized by selecting the correct bait, layout, seasonal timing, and length of operation to suit the purpose (King et al. 1994; O'Donnell et al. 1996), trapping is always labor intensive, and therefore costly.

Take, for example, the effect of trap spacing on the sex ratio of the catch. In an early experiment setting out an array of kill traps at different spacings, the wider the distance between them, the fewer stoats were caught per kilometer of trap line and the higher the proportion of males (Table 13.2). The removal of even large numbers of males makes little difference to the number of young born the next spring, since stoats are polygynous and males are very efficient at finding estrous females (Chapter 9). All females are already pregnant by midsummer, including the young females that are fertilized before they leave the den, so any that escape trapping can produce a litter even if every male in the area is killed. Put another way, wide-spaced traps may do more harm than good, because they remove mostly males. Removing males leaves more food for the surviving females, and may even enhance the ability of the local stoat population to recover. Therefore, although a widely spaced array of traps may be less effort to operate (Chapter 8), it will have little effect on the next year's population if it does not remove enough females.

Research on improving baits and lures has been intense in recent years. The most effective lures for stoats are fresh meats or eggs, but many new baits are being tested (Spurr 1999; Montague 2002; Spurr et al. 2002). In response to field reports of weasels hunting by sound (Willey 1970), electronically produced sound lures have been tried, but with equivocal results (Spurr & O'Connor 1999). Synthetic lures incorporating components of stoat musk, a natural scent marker of great significance to stoats (Chapter 8), have been, so far, not very successful (Clapperton et al. 1994). Natural lures made from real anal scent glands help (Clapperton et al. 1999), but obtaining large supplies of the natural product would be logistically difficult.

All other artificial odors and flavors offered to stoats in pen trials (e.g., commercial trappers' lures, food flavors, trimethylamine, and synthetic fermented

Table 13.2 The Effect of the Spacing Between Kill Traps on the Number of Male and Female Stoats Caught Over 17 Months

Distance between traps, m

Table 13.2 The Effect of the Spacing Between Kill Traps on the Number of Male and Female Stoats Caught Over 17 Months

Distance between traps, m






Number of traps






Length of line (km)






Total number of stoats caught






Sex ratio (% males)






Stoats caught per 100 trap nights






Stoats caught per km of trapline






(Data from experimental studies in southern New Zealand in 1972-1976, reported by King 1980a.)

(Data from experimental studies in southern New Zealand in 1972-1976, reported by King 1980a.)

egg) were unattractive (Spurr 1999). Perhaps young predators learn what prey are worth hunting and safe to eat during a critical period of their development, so they might not recognize artificial baits as potential food. If so, the best baits to offer may depend on the natural diet of stoats in each location. Bait markers help in field trials of new baits, because they can identify which animals ate the bait and which did not. Markers such as rhodamine B and iophenoxic acid are tasteless and can be detected in the whiskers and blood, respectively, of stoats that had eaten marked bait up to several weeks previously (Spurr 2002a, 2002b; Purdey et al. 2003).

One early set of field trials showed that capture rates of stoats were higher early in a trapping session than later, and higher in baited than in unbaited traps (King 1980a). The most efficient way to run a trap line in that situation was, therefore, to set the traps close together, bait them well, inspect them daily for a few days, and then spring them closed and do something else for the rest of the month.

In that trial, the number of stoats caught fell by about 60% by the end of each 14-day session, although immigrants soon replaced those caught by the beginning of the next month's trapping. It seemed likely that less than half the number of stoats present in the forest had been caught, even though the greatest practicable effort had been expended. Ratz (1997) calculated a linear regression comparing capture rate with cumulative number of stoats removed, which gave median estimates (with very wide confidence intervals) of 58% of stoats removed after 8 days and 87% after 12 days. Unfortunately, it takes very few stoats to make a lot of difference to a local population of threatened birds, so the risk remains that removal rates like these may not be sufficient to achieve a useful level of protection.

In an unstable population, a decline in the numbers of stoats after all that effort is not necessarily proof of the effectiveness of control, because trapping often does not add to, rather than replace, natural mortality (McDonald & Harris 2002). Moreover, young stoats are able to travel very long distances (20 km or more) within a few weeks of independence (Chapter 9), and can repopulate cleared areas within 2 to 3 months (Murphy & Dowding 1994; Alterio 1996; Murphy et al. 1999).

On the main islands of New Zealand, therefore, the intensity of trapping required to remove more stoats than die from natural causes necessarily involves a huge commitment of labor for a temporary result. The Department of Conservation does not conclude that stoat trapping is worthless; rather, they see it as "rather like cleaning toilets, you just have to keep on doing it" (E.C. Murphy, personal communication). But the incentive to find more cost-effective methods is very, very strong.

Only on the offshore islands can rapid recolonization by stoats be prevented, and then only if they are isolated by more than about 2 km of water. The story of Maud Island (309 ha), an important rodent-free sanctuary in the Marlborough Sounds (at the northern end of the South Island), illustrates two of the most extraordinary abilities of stoats: swimming and evading traps.

Crouchley (1994) described how Maud has been invaded three times in a decade by stoats swimming across the 900-m salt-water channel from the nearest mainland. In April 1982, the first stoat was discovered because it responded to taped calls of the saddleback, an endemic forest bird that had recently been introduced to the island. Traps (up to 80) were immediately set all over the island, but it was 7 months before the first of seven young stoats was caught, and it took 16 months to get the last of them. The first one seen must have been a female, carrying (as usual) a litter of unimplanted blastocysts (Chapter 9), and the seven stoats captured were presumably the young she produced on the island. She herself was never caught.

A similar story was repeated on the same island in 1989, and this time the litter included at least five young. Like the previous immigrant adult female, this one was also never trapped despite the intense efforts of the Maud Island staff. She lived on the island for 18 months until she was found dead in February 1991. By then, sibling matings between her offspring had begun to produce the next generation, of whom at least three were included among the total of16 stoats caught up to July 1994. In March 2003 a stoat was discovered on the island for the third time—fortunately, this time it was a young male. Obviously, small islands lying as close to the main islands as Maud will never be safe from these repeated incursions.

By contrast, on Te Kakahu (Chalky Island, 514 ha) off the southwest coast of Fiordland, stoats had lived for many years free of all contact with humans. The island supported no rodents, but several large breeding colonies of burrowing seabirds and of New Zealand fur seals provided plenty of eggs, chicks, and carrion from early spring to late autumn. During the winter their primary diet was bush birds, supplemented by carcasses of adult seabirds and perhaps also cached chicks and eggs. Like all stoats, however, they no doubt still preferred fresh meat if they could get it.

On that assumption, the Department of Conservation planned an intensive trapping campaign for the winter of 1999 (Willans 2000). The plan was to avoid the Maud experience by timing the start of trapping just at the time that resident stoats had few live prey to hunt, and by making sure that every stoat had access to several traps and could not learn to avoid them until too late. After the last of the young seabirds had fledged in April and May of 1999, 140 double trap tunnels were placed at 100-m intervals all over the island and prebaited with fresh eggs and day-old chicks for 2 weeks. Most of these were quickly taken, presumably by hungry stoats.

When the traps were set on June 29, 15 stoats were killed in the first 2 weeks (11 on the first night!), and one more was found when the team next visited the traps in October. Further offerings of fresh baits remained untouched, trained stoat dogs could find nothing, and stoat footprints were no longer seen on the sandy beaches of the island. Te Kakahu is now available for restocking with threatened species from the mainland, and the numbers of small bush birds are recovering. The once-only cost of removing the stoats (NZ $62,000 over 14 months to June 2000; M. Willans personal communication) is trivial beside the present and future value of the island as a predator-free reserve.

It is beyond belief that the two female stoats that invaded Maud could have avoided traps for months, when the whole resident population on Te Kakahu was eradicated within weeks. Te Kakahu was not a fluke, either: Similar operations have been repeated on other Fiordland islands (in Anchor Island, 2001, and Bauza Island in 2002) (M. Willans, unpublished). The contrast between these stories illustrates how well stoats can learn to avoid traps (Chapter 8). Unfortunately, textbook operations like the ones in Fiordland are often logistically impossible to repeat on the main islands, but they certainly add to other pressures to find means other than trapping for protecting native fauna from stoats.


New Zealand legislation controls but does not prohibit primary poisoning programs targeted directly at stoats. No toxins are presently registered for use against stoats, although some (including 1080, diphacinone, and cholecalcif-erol) may be used under experimental research permits; trials so far show that all have drawbacks (Lawrence & Dilks 2000; Spurr 2000; Spurr et al. 2002). Poisoning is unpleasant to think about, but the looming extinction of irreplaceable endemic birds is worse. Consequently, in New Zealand toxins have remained a permissible tool for predator control long after most other countries have banned them. The toxins used, the research programs testing them, and the people handling the most dangerous of them, are all subject to stringent supervision by regulatory agencies concerned as much to ensure humane treatment of the pest animals as to protect nontarget species and human interests.

One toxin commonly used is 1080 (sodium monofluroacetate), a natural plant toxin widespread in native vegetation in Western Australia, which is very effective and breaks down quickly in soil and natural water. Stoats are very susceptible to minute amounts of 1080 (1 mg is a fatal dose for stoats of all sizes), and it acts quickly and apparently humanely (Potter & King unpublished). When good delivery methods have been developed that can fully protect nontarget species, 1080 could become a viable alternative to trapping for targeted use against stoats.

Rodenticides are widely used to control rats and voles on farms worldwide (Chapter 12), and stoats are vulnerable to secondary poisoning when they eat dead or sublethally poisoned rodents (McDonald et al. 1998; Shore et al. 1999). In New Zealand, where the only native mammals that could be affected are rare, semiterrestrial short-tailed bats (King 2005a), 1080 has been an important and routine weapon against rabbits, rats, and possums for several decades.

Secondary poisoning has probably had local and temporary effects on stoat populations for much of that time (Gillies & Pierce 1999; Murphy et al. 1999; Alterio 2000). Since most introduced mammals are regarded as undesirable in New Zealand, secondary poisoning is less a cause for concern to managers than it is elsewhere. The argument is that the ecological costs of using toxins against introduced mammals are much less than the damage costs to native fauna if they are not used, although we still need to learn more about the ecosystem-level consequences (Innes & Barker 1999).

Nonlethal Control Methods

Arguments about the ethics and principles of lethal predator control and its use in conservation management have accelerated in recent years, and seem likely to continue (Littin et al. 2004). In this context, animal welfare is concerned with individual pests, whereas conservation is concerned with protection of populations in the wild (Eggleston et al. 2003). Both concepts are important but, unfortunately, sometimes incompatible. If we accept that all animals have the same rights to protection without consideration of the damage some may do to others, then we are ignoring our responsibility to vulnerable species and to biodiversity generally (Fulton & Ford 2001).

It could be argued that lethal control of exotic predators is justifiable to protect threatened species, especially species that are disadvantaged by human activities such as intensive agriculture and chemical sprays. Effective eradication using humane lethal methods not only is cost-effective, but also makes all further killing unnecessary. This process is well advanced, especially on the offshore islands of New Zealand (Towns et al. 1997; Towns & Broome 2003). Islands will always be the most important refuges, and progress in clearing stoats and rodents from all suitable offshore islands continues apace (Parkes & Murphy 2003). An alternative strategy would be to develop nonlethal means of protecting vulnerable prey.

Nonlethal control is possible if the aim is to prevent damage rather than reduce predator numbers. For example, an obvious solution to protect nesting game birds is a predator-proof fence. In North America, several designs for electrified predator-proof fences kept out predators of the size of foxes, skunks, and raccoons, but not stoats or raptors. The cost per additional successful nest was over $100 in two cases reviewed by Reynolds and Tapper (1996), and the consequences of even a single fox breaching the fence were so catastrophic that predator removal continues to be the preferred policy.

New Zealand has a smaller suite of predators and, most significantly, no foxes. There, continued research has produced a genuinely practical nonelectri-fied design effective against stoats (Clapperton & Day 2001), and even mice. New Zealand is leading the world in the technology of creating predator-proof enclosures. At the time of writing (2006), some 15 of these enclosures have been built around New Zealand, and another five are imminent, at massive cost but with enthusiastic community support (R. McGibbon, personal communication). Some of the earliest enclosures already support protected species. Other nonlethal methods, such as repellents (Spurr 1997) and fertility control (Cowan & Tyndale-Biscoe 1997; Norbury 2000), are also being developed. Potential new techniques involving genetic engineering of vectors to carry stoat-specific bio-control agents is not ruled out, but would require extensive public consultation (Fitzgerald et al. 2002), and in general a better strategy is to improve current methods first.

Was this article helpful?

0 0

Post a comment