Demographic resilience of territorial island birds to extinction: the flightless Aldabra Rail Dryolimnas (cuvieri) aldabranus as an example

Abstract

Since 1600, a disproportionate number of avian extinctions have occurred among flightless and island-dwelling species. Some of these happened very rapidly, implying that such populations had low resilience to perturbation. In managing insular populations, there is a need to be able to predict their demographic responses to novel circumstances, such as predator introduction, not least in order to quantify the window of time within which remedial action must be taken to minimise extinction risk. To explore how such resilience might be quantified, we used the territorial, flightless Aldabra Rail Dryolimnas (cuvieri) aldabranus as a ‘model’ species. Endemic to Aldabra Atoll, Republic of Seychelles, this is the last remaining flightless bird in the tropical Indian Ocean. Formerly ubiquitous on Aldabra, by 1977 its range had contracted to three islands: Malabar (by farthe largest population), Polymnie and Île aux Cèdres. In 1999, the species was successfully reintroduced to a fourth island (Picard), where exponential growth of the reintroduced population was predicted to continue until it reached c. 1 000 pairs in about 2010. Despite this success, the entire world population of Aldabra Rails remains confined to four adjacent islets, with no ex situ populations, placing the species at high risk of stochastic extinction. The aims of this study were to assess whether there were any significant changes in the population size of rails in recent decades and to explore the population-level consequences of the most likely stochastic event, viz. the introduction of alien predators to Malabar Island (cats being abundant on the adjacent island of Grande Terre). There have been no substantial changes in the population at Malabar Island since the 1980s, but the previous estimate of the number of breeding pairs was revised downwards from c. 4 000 to c. 3 500. We made the first assessment of the number of non-breeding birds (floaters)—in early 2000, Malabar was estimated to have c. 3 500 floaters—and explored their role as a demographic buffer to extinction using a stochastic population model. Despite the large number of floaters, a population of 60 or more cats is predicted to drive the Malabar rails to extinction within 20 years. However, in the absence of introduced predators the Malabar population is resilient to the removal of up to 100 pairs of breeding birds annually for introductions to other islands within the species’ former range. Of wider relevance, for territorial, insular species, monitoring of the floater population may provide an earlier warning of a pending population crash than monitoring of numbers breeding.