The effects of lack of food as a cause of mortality are complicated because starvation acts together with other environmental stresses. It is rare to find any animal dying just from starvation, as usually some other, more immediate, cause can also be identified and labelled as the culprit. In England and Germany, starvation was deemed the sole cause of death in none (Simpson 1997) and only 1.3% (Hauer et al. 2002a) of inspected otter carcasses. However, the role of starvation may show in different ways. For instance, if food shortage were involved as a major factor causing mortality, one might expect more otters to die (from whatever proximate cause) at times of low food availability, and with associated low body condition.
In some high-density areas, mortality of Eurasian otters, other than by violent death, is indeed highly seasonal (Kruuk and Conroy 1991; Kruuk et al. 1987). In Shetland we found most 'natural' deaths in spring, with 60% of all natural deaths occurring between March and June. This was a significant seasonality, in sharp contrast with the even distribution of violent deaths throughout the year (Fig. 12.12). Similarly, we reported that, on the Scottish mainland, 42% of non-violent deaths occurred in April alone—more than in any other month. Violent deaths, such as 73 cases of road death, were evenly distributed over the months (12% in April), a significant difference (Kruuk et al. 1993a). The seasonality is paralleled by the availability of important prey species in both sea and fresh water: fish populations are at their lowest in early spring (see Chapter 8; Kruuk et al. 1987, 1988, 1993a). Furthermore, low prey availability occurs at a time of year when the otters' metabolic requirements are high, owing to low water temperatures (see Chapters 9 and 10).
/ Non-violent mortality
/ Non-violent mortality
Figure 12.12 Death of Eurasian otters from violent and non-violent causes in Shetland. Mortality from non-violent causes (n = 49) was seasonal, peaking in late spring (x2 = 20.1, 5 d.f., P < 0.01). There was no significant seasonal variation in violent deaths (n = 53). (After Kruuk and Conroy 1991.)
Despite such evidence of seasonal starvation, we found no seasonality in the body condition of otters: animals killed on roads or by other violent means showed the same relative bodyweight throughout the year (Kruuk and Conroy 1991). However, when an animal dies 'naturally' in spring, its bodyweight is significantly lower. The likely explanation for this is that otters have only small fat reserves at any time of year, and that when an animal starts losing condition (i.e. starts using those reserves) it is likely to die quickly.
If starvation were an important mechanism of population limitation in otters, one would expect the animals to take a large proportion of the available prey population. In addition, one would predict that during the time of low prey availability otters would switch to suboptimal types of prey such as carrion, birds and mammals. Both predictions have been confirmed in rivers and streams of mainland Scotland (see Chapters 7 and 8).
The presence (numbers and time spent) of Eurasian otters in Scottish inland rivers and streams (see Chapter 5) was significantly correlated with fish biomass, and hence productivity, although the sample was small (Kruuk etal. 1993a) (Fig. 12.13). Similarly, absence of otters in southern Norway was explained by the decline of fish populations due to acid rain (Heggberget and Myrberget 1979), and in Spain a positive correlation was found between otter density and fish numbers (Ruiz-Olmo et al. 2001). With all the various strands of evidence, it seems likely, therefore, that at least in some areas prey shortage was an important ultimate factor determining otter
numbers. It may affect mortality, as suggested above, and possibly also reproduction (see Chapter 11).
There is evidence that the upper limits of sea otter numbers may be determined by the 'carrying capacity' of the habitat, and the interaction makes for a fascinating story. In the Aleutian islands Attu and Amchitka, Estes (1990) demonstrated that in areas where sea otters are newly arriving and prey (especially sea urchins, shellfish and crabs) is plentiful, populations increase through reproduction at their maximum rate, 20-24% per year. Once populations are at 'equilibrium density' (the habitat's 'carrying capacity'), mortality from starvation increases (Kenyon 1969), especially amongst young otters. Numbers do not increase any further. After some years at this density, the sea otters have to spend up to three times as long every day to obtain sufficient food, as prey numbers decrease owing to high predation by otters.
Next, because of the otter-induced reduction in sea urchin numbers, sea urchin grazing of kelp is reduced and, where sea urchin barrens existed previously, kelp forests are able to establish. This sets the scene for a quite different ecosystem, with substantial populations of fish in the dense stands of kelp. Sea otter diet changes from invertebrates to fish. The kelp forests have a significantly higher carrying capacity for sea otters, and another, higher, 'equilibrium density' is established.
This intriguing scenario was confirmed and amplified by Monson et al. (2000), comparing populations between Amchitka and Kodiak islands before the recent decline set in (see below). There were equal birth rates in the two areas, but in the equilibrium population at Amchitka the percentage of pups being weaned was about half of that in the growing population at Kodiak (52% versus 94%). The rest died mostly within the first month of life. Adult mortality rates were similar in the two areas. A similar comparison was made by Monnett and Rotterman (2000), between sea otters in Kodiak and Prince William Sound. Until the early 1990s, early pup survival was probably the primary factor regulating sea otter populations.
Was this article helpful?