The uncertainty paradox of SLOSS

The SLOSS debate has been unresolved for decades in conservation biology, ever since Jared Diamond asked whether a Single Large Or Several Small reserves maximized the conservation of species. Further research, including some of my own, has shown that the answer to this question depends on the extinction risk of species, with very many small reserves being optimal when extinction risk is low, and single reserves being optimal when extinction risk is high.

But estimates of extinction risk are routinely uncertain. Given this, you might think that this uncertainty would cloud the choice of the optimal design of nature reserves. However, if the problem is re-cast, and we seek a reserve system that is most robust to uncertainty in the risk of extinction, we find that a small number of reserves is optimal, and that a large number of reserves is risky.

This seems paradoxical. How can accounting for uncertainty in one of the key parameters actually help resolve the debate? To find out more, please read a Decision Point article about my recent research on the logic of reserve design, or the original scientific article in Ecology Letters. Please email me for a copy (this should open an already-written request in your email client, and send an auto-reply).

Fig. 1. The number of reserves that maximizes the probability that the risk of extinction of species in a reserve system is less than 1/a for different values of b. The parameter b controls how the risk of extinction within a single patch scales with patch area. Thus, the optimal number of reserves is driven largely by the aspiration (a) for the extinction risk that is sought by a manager rather than the population dynamics of the species. In fact, the actual extinction risk, or even any description of the uncertainty in this risk does not influence the result at all.

This figure is reproduced from the original article in Ecology Letters (McCarthy, M.A., Thompson, C.J., Moore, A.L, and Possingham, H.P. 2011. Designing nature reserves in the face of uncertainty. Ecology Letters 14: 470-475).

About Michael McCarthy

I conduct research on environmental decision making and quantitative ecology. My teaching is mainly at post-grad level at The University of Melbourne.
This entry was posted in Ecological models, New research, Probability and Bayesian analysis and tagged , , , . Bookmark the permalink.