Alarm Calling and Predator Awareness Training in the Critically Endangered Helmeted Honeyeater
Abstract
Captive breeding and reintroduction are key to modern conservation, but high predation of released animals means reintroductions often fail. The captive environment can provide individuals with little opportunity to learn about predators or alarm calls. As a consequence, captive-raised animals released into the wild remain naive to predators, commonly suffering catastrophic predation. This issue is a severe challenge for modern conservation, which is increasingly reliant on captive-breeding and release to establish or reinforce endangered populations. As a result of this ongoing challenge, conservation research is now examining approaches to addressing naivety in captive-bred individuals prior to release. One approach to addressing loss of anti-predator behaviour is predator-awareness training, where captive-bred individuals are exposed to predators or predator cues prior to release to help them develop appropriate anti-predator behaviours. This principal has been well examined in fish, where individuals rapidly associate novel predators with known chemical alarm cues. However, results of programs in terrestrial vertebrates have been much more mixed; the few programs that have been conducted vary broadly in methods, and while some have been successful in altering behaviour of captive populations, success in improving post-release mortality has been far less clear. Given the success of predator-awareness training in aquatic systems, training in terrestrial vertebrates deserves more attention. In this thesis I examine alarm calling and predator-awareness training in the critically endangered helmeted honeyeater, Lichenostomus melanops cassidix. Since the 1980s the helmeted honeyeater has been the focus of an intensive captive breeding and release program, however release success has been challenged by high post-release predation by native hawks. The history of successful captive-breeding, and ongoing challenges to release success in this captive population provides an excellent opportunity to develop a new training program, and to address questions around alarm calling behaviour in captivity. The thesis has six chapters, including an introduction, four data chapters, and a synthesis. Chapter 2 provides a comprehensive review of predator-awareness training programs in terrestrial vertebrates. I survey published and unpublished studies to examine methodological approaches in past training programs, and determine how these approaches have affected success. The results of this review inform the structure and focus the later chapters. Chapter 2 has been published as a paper in Biological Conservation. In Chapter 3, I use playback experiments to examine the response of wild-living honeyeaters to conspecific and heterospecific alarm calls. I describe anti-predator responses to alarm call playbacks, and use these results as a baseline for functional, natural anti-predator behaviours in this species. The results of this chapter allow comparison of the behaviour of captive birds with those of successful individuals in the wild. Chapter 4 describes a predator-awareness training program. I examine if response of captive juveniles to alarm calls differs from that of wild-living birds, and whether training has an effect on these responses in captivity. Finally, I examine if training improves post-release mortality by comparing survival of trained and untrained cohorts following release into the wild. Chapter 5 addresses indirectly the effect of captivity on predator recognition in captive helmeted honeyeaters, by examining heterospecific response to honeyeater alarms. I use playback experiments examine how heterospecifics living near and far from captive honeyeaters respond to honeyeater alarms. If honeyeater calls reliably indicate danger, local birds should learn to associate honeyeater calls with danger and respond accordingly, whereas distant birds should not respond to playback of the unfamiliar alarms.