Stoats in New Zealand Beech Forests

In New Zealand, the common land mammals in the forests are few and they all belong to introduced species (Chapter 5). Stoats are usually the most common, and often the only, mammalian carnivores. Native raptors (the Australasian harrier hawk, the bush falcon and the morepork, a woodland owl) are scarce. This animal community is not "natural," but at least it is very simple. That makes it easier to see what is happening, and the sequence of events is seldom complicated by competition at any level. One can hardly imagine circumstances for stoats to live in (habitat, prey resources, competitors) more radically different from those in Sweden. Contrasts like these provide exciting opportunities to understand the population dynamics of these adaptable little predators.

King (1983b) was interested in documenting the responses of the forest rodents and mustelids to the masting cycle of the southern beech trees, especially in two mountain valleys in Fiordland National Park, in the far south of the South Island (Figure 10.4). Every 3 to 4 years the beech trees flower synchronously, and when the spent beech flowers fall, the litter-feeding invertebrates get a boost in food supplies (Alley et al. 2001). Several months later, a massive seedfall dumps tons of food onto the forest floor in autumn and early winter (March to April).

The effects are rather like those of a stone dropped in a pond. In both forests, 1976 and 1979 were masting years. Mice benefited from both the additional in-

Figure 10.4 The variation from year to year in the numbers of feral house mice and of stoats in southern beech forests in New Zealand. The year in which the beech trees produced a heavy crop of seed is denoted as Year 1. (After King 2002.)

vertebrates and from the seeds. Over the few months following seedfall, the numbers of young mice caught on standard traplines soared, not only because more were born but also because more survived to enter the traps. Further, the adult female mice continued to breed well into the winter, instead of stopping in autumn as usual. By the southern spring (September), an extra large group of young adult mice was already breeding and producing the first of the summer generations. By early summer (November), the mice had reached the relatively high population densities (for the season) of up to 20 mice per 100 TN. The response of rats to the seedfall was modest during the 1970s (King & Moller 1997), but more obvious after a double seedfall in 1999 and 2000 (Dilks et al. 2003).

For the stoats living in the same forests, life was suddenly much easier than usual. Stoats benefited both from the extra mice and from the increased numbers of seed-eating birds (Murphy & Dowding 1995). For the next few months they feasted. Stoats of both sexes and all ages ate mice much more often than usual. But, more important, the increase in the supply of mice in these forests affected the stoats out of all proportion to the increase of mice in the stoats' diet (King 2002; King et al. 2003b).

In nonseed years the density index for stoats ranged from 1 to 2 C per 100 TN in summer down to zero in winter. But in the summer after a big seedfall, the density index for stoats shot up to 5 to 6 C per 100 TN. The increase in numbers of stoats caught in postseedfall summers was directly related to the density index for mice at the same time. The whole chain of events took less than a year, from the March, April, and May of the main seedfall to the following December and January when the season's crop of young stoats dispersed.

The sequence was originally worked out from a large sample of New Zealand stoat carcasses collected in the 1970s (King & Moody 1982; King 1983b; Powell & King 1997; King 2002), and confirmed in the same or similar areas many times since (Murphy & Dowding 1995; O'Donnell & Phillipson 1996; Wilson et al. 1998; King et al. 2003b; Purdey et al. 2004). The data have been used to construct three independent computer models of the relationships between stoats and rodents in this simple, feast-or-famine environment (Blackwell et al. 2001; Barlow & Choquenot 2002; Wittmer et al. unpublished).

The number of young that can be produced by any population of animals in a given season depends on four things: (1) the number of females in breeding condition; (2) their fecundity, or the mean ovulation rate per female; (3) their fertility, or the mean litter size per female; and (4) their productivity, or the number of young reared to independence over the whole local population. For stoats, the number of females in breeding condition can be taken as 100% every year, since almost all females of all ages are fertilized by the end of each breeding season (Chapter 9). We can therefore ignore the first point, but the others are all important.

Fecundity is easy to measure, by counting the corpora lutea in the ovaries or the blastocysts in fresh uteri. By contrast, fertility is very difficult to measure in stoats, because it is impossible to collect large samples of pregnant females or, until very recently, to find breeding dens to observe (see p. 255). Productivity is also very variable, but easily deduced from the age structure of the summer population. So we have the beginning and the end of the story, but there is a gap in the middle. Still, we can see enough to work out the general outline.

Environmental conditions acting on female stoats control the number of young produced each season by a simple, energy-saving, and effective mechanism. The females cannot increase their fecundity in a good year, as common weasels do. The potential number of young stoats born in any given year is already set by the number of ova shed in the previous year, and delayed implantation fixes the cycle regardless of food supplies. The female stoats have no opportunity to increase that number even in a bonanza season. The only thing that can happen is a decrease in the mortality of the young at all stages of their lives, from implantation of the blastocysts to independence.

When food is short, some blastocysts fail to implant, or some of the embryos are resorbed before reaching full term, or some of the young born die as nestlings; eventually these losses, or some combination of all of them, reduce litter size to a manageable number. When food is abundant, few potential young are lost at each stage. To follow the process, we need to calculate how many young start off each year as ova, and then how many fall aside at implantation, in the uterus, at birth, in the den, and during the transition to independence.

Was this article helpful?

0 0

Post a comment