Endangered metapopulations and disease

As discussed previously, metapopulation ecology has come to the fore as a theoretical framework for conservation planning. High dispersal rates (high movement between patches) have been predicted to increase the proportion of patches occupied at equilibrium, the time of metapopulation extinction, and the effective population size (Hanski and Gilpin 1997). Conservationists have favored measures, such as habitat corridors, which increase connectivity among patches. Hess (1996), however, has suggested that easy migration can have the negative effect of spreading disease among patches, causing extensive local extinctions. Using a metapopulation analysis, Hess found that high migration rates, by facilitating the movement of disease organisms, could reduce patch occupancy and increase the probability of metapopulation extinction.

Gog et al. (2002), however, disagreed. They believe that most of the infections that threaten wildlife are not caused by migration of diseased organisms, but by "spillover" from other, more abundant hosts already present in the habitat patches. The reservoir for these "spillover" diseases is often domestic animals. For example, domestic dogs are the probable source of diseases that have threatened African wild dogs (Lycaon pictus), African lions (Panthera leo), Baikal seals (Phoca sibirica), grey wolves (Canis lupus), and arctic foxes (Alopex lagopus semenovi) (Gog et al. 2002).

In the deterministic model of Gog et al. (2002), S is the proportion of susceptible host patches (host population present, no disease), and I is the proportion of infected patches (host population and disease present). The extinction rates of susceptible and infected populations are xS and xI, respectively. The migration rate between susceptible and infected populations is y. When an infected disperser arrives at a susceptible patch, it infects the resident population with the probability of S. Infection spreads at the rate of ySIS. The preceding is identical to the Hess (1996) model. What Gog et al. (2002) added is an extension of the Hess model in which they simulated various parameters of an infection rate from an "outside source," g. Starting with the Hess model, Gog et al. set g at zero, then ran a number of simulations showing the important effects when g is a non-zero parameter.

The equations for mean proportion of patches occupied in the S and I states are:

Equations for stable equilibrium values of S and I for different values of g and m are found in appendix A of Gog et al. (2002). Figure 9.1 has been produced based on their equations for g = 0 (representing no outside sources of disease) and g = 0.4 (representing a moderately large background infection rate). Other parameters are the same as in Hess (1996): xs = 1.4, xI = 2.4, S= 0.5.

As we see from Fig. 9.1, Gog et al. (2002) found that when g is zero or very small (as in Hess), occupancy rates first increase but then decrease with increased migration as more and more patches experience extinction due to disease. Eventually, patch occupancy increases again with more migration, as all patches become infected. This result (Fig. 9.1a) led Hess (1996) and others to suggest that increased migration between patches can have a negative effect on patch occupancy and can increase the probability of metapopulation

Movement rate, m

Figure 9.1 Proportion of suitable patches occupied as a function of movement rate: (a) with the parameter g = 0; (b) with g = 0.4. When g = 0.4 there is a reasonably large chance of infection from an "outside source." Adapted from Gog et al. (2002) and Hess (1996).

Movement rate, m

Figure 9.1 Proportion of suitable patches occupied as a function of movement rate: (a) with the parameter g = 0; (b) with g = 0.4. When g = 0.4 there is a reasonably large chance of infection from an "outside source." Adapted from Gog et al. (2002) and Hess (1996).

extinction. However, for larger values of g, increasing the migration rate results in little if any depression in the rate of patch occupancy. In other words, the decrease in patch occupancy at intermediate levels of migration is minimized (Fig. 9.1b). Gog et al. (2002) concluded that the net effect of migration is almost always positive, and that at high rates of infection from external sources the benefits of migration will always outweigh the costs. The major application of these models is that in wild populations suffering from a high rate of infection from alternative host species, patch occupancy should increase, rather the decrease, with migration rate.

Gog et al. (2002) concluded that corridors between suitable habitats are likely to benefit metapopulation persistence, a conclusion also reached by Laurance and Laurance (2003; see Chapter 5). However, they pointed out that the Hess model might apply well to captive populations, where transfer of animals from one facility to another is often a cause of disease epidemics. They stressed the importance of veterinary screening and quarantining procedures before transferring animals from one captive population to another.

Was this article helpful?

0 0

Post a comment