Continental-level biodiversity collapse.

Biodiversity loss is a global phenomenon (1), with Woinarski et al. (2) providing a sobering review of the demise of the Australian native terrestrial mammal fauna. The authors’ focus is on the loss of >10% of endemic terrestrial native mammal fauna since European settlement in 1788. The extent of mammal loss is highlighted by the contrast with mainland North America, with just one mammal species extinction in the equivalent ∼200-y period. Woinarski et al. (2) make key points about conservation policies and practices, some of which have global implications. Woinarski et al. (2) provide convincing evidence for predation by the feral cat (Felis cattus) and red fox (Vulpes vulpes) as a primary driver of mammal decline (Fig. 1A). The authors also contend that altered fire regimes, in part through the loss of indigenous land custodians, are an important contributor to Australian mammal decline, particularly through promoting hunting efficiency by feral predators. Direct interventions must be taken to safeguard the most imperiled Australian mammal species. New approaches within unfenced areas, such as improved poison baits, are urgently required to control feral animals (particularly the feral cat). An expanded network of feral predator-free fenced areas is needed on the Australian mainland. Woinarski et al. (2) discuss how such initiatives can be highly successful. However, they can sometimes be so successful that other problems arise, such as overabundant populations of prey species (e.g., macropods) (3) and highly unexpected cascading extinctions of native mammal species of conservation concern (4). For example, intensive poison baiting in Booderee National Park in southeastern Australia resulted in significant suppression of red fox populations. However, at the same time, the treedwelling marsupial greater glider (Petauroides volans) (Fig. 1B) became regionally extinct and others, including the common ringtail possum (Pesudocheirus peregrinus), are in steep decline (4). Such “ecological surprises” underscore the critical need for careful monitoring of management interventions, including feral predator control. Although concerted action is essential to reduce populations of feral animals in many parts of Australia, it will be critical not to lose sight of the need to tackle other key drivers of biodiversity decline. This is because predation by feral animals (and its interaction with altered fire regimes) is not the primary driver of biodiversity decline (including mammal decline) in some Australian ecosystems. Contextspecific actions are therefore required to address context-specific threatening processes for particular species and ecosystems of concern. For example, industrial logging, recurrent wildfire, and long-term decline of both large old trees and stands of old-growth forest, are the wellestablished interacting drivers of decline of the critically endangered Leadbeater’s possum in the wet forests of Victoria (5). Cessation of industrial logging is needed to prevent the extinction of this species. Moreover, a return to indigenous burning regimes is not a valid strategy in these forests, because there is little evidence of pre-European Aboriginal presence in these ecosystems. Australia’s temperate woodlands provide another example. There, biodiversity decline is a result of a suite of interacting factors beyond the effects of exotic predators, such as long-term vegetation clearing [including burning of large old trees (Fig. 1D) and firewood harvesting], overgrazing by domestic livestock, and past intensive hunting and bounty programs (6). Large-scale and long-term replanting and natural regeneration programs, as well as better protection of large old trees, will be critical for restoring both temperate woodlands and many elements of the biota they support, including some mammals of conservation concern (7) and threatened bird species (8). Woinarski et al. (2) only briefly touch on range collapse in extant Australian mammal fauna. Some species have been lost from 95– 99% of their former distribution (9), with range contractions so profound that many people remain unaware of the array of native mammals that used to occupy certain areas: the widely discussed shifting-baselines concept (10). An example is the greater bilby (Macrotis lagotis) (Fig. 1C), an animal now associated with remote arid environments. Many Australians, including some conservation scientists, are shocked to learn that the species was once common and widely distributed across mesic environments, like woodland ecosystems close to major urban centers, such as Canberra. The local and regional extinction of this species (and a suite of other medium-sized native mammals lost from temperate woodlands) has led to the functional extinction of the key ecosystem