Determinants of resource specialisation and its ecological consequences for the corallivorous filefish, Oxymonacanthus longirostris

All organisms are dependent upon their habitat for the resources necessary for survival and reproduction, but they vary immensely in the degree to which they are specialized on particular resources. Habitat specialists appear to evolve in environments where preferred resources are stable, however specialists are vulnerable to extinction should these resources become scarce. Given global declines in the quantity and quality of available habitats, it is critical to understand why species are specialised, and what resources they are specialized on, in order to determine their vulnerability to habitat loss. Coral reefs are home to a diverse assemblage of coral-associated organisms and of these, corallivorous fishes are among the most specialised. One such species, the harlequin filefish, Oxymonacanthus longirostris, is distributed throughout the Indo-Pacific in areas of high coral cover. This study investigated the ecology of O. longirostris to determine how specialised it is, the roles of prey quality and predation pressure as drivers of specialisation, and the consequences of being a coral specialist in light of increasing coral degradation. Most animals consume a narrower range of food resources than is potentially available in the environment. Chapter 2 examined the drivers of this prey selection by testing the consequences of prey preferences for key fitness-related parameters. While the diet of O. longirostris was dominated by the most abundant coral species, Acropora nobilis, fish appeared to preferentially feed on rarer acroporids, such as Acropora millepora. Choice experiments confirmed strong preferences for these rarer corals, suggesting diet is constrained by availability of different coral species. In a feeding experiment, reproductive pairs fed on non-preferred corals exhibited dramatic declines in condition and reproductive output compared with those fed preferred corals. These experiments suggest that fish distinguish between available corals based on their intrinsic value as prey, that reproductive success is dependent on the presence of particular coral species, and that differential loss of preferred corals could have significant consequences for population success. Foraging theory predicts that preferred prey should maximise energetic reward relative to the energy expended to access, capture, and consume prey. However, the relative roles of differences in the nutritive value of prey and costs associated with differences in prey accessibility are not always clear. Chapter 3 investigated within-colony feeding in O. longirostris to establish if prey accessibility determines foraging patterns. O. longirostris exhibited non-uniform patterns of foraging in the field, feeding midway along branches. On simulated corals, fish replicated this pattern when food accessibility was equal along the branch. However, when food access varied, fish consistently modified their foraging behaviour, feeding where food was most accessible. When foraging patterns were compared with coral morphology, fish preferred larger polyps and less skeletal protection. Consequently, it appears that patterns of prey selectivity are influenced by coral morphology, with fish preferring corals with structural characteristics that increase prey accessibility. A selective coral diet may further benefit O. longirostris by reducing predation risk. Many organisms primarily interact with their surroundings using non-visual sensory systems and may have evolved mechanisms to 'blend in' with chemical components of their habitat. One potential mechanism of camouflage is through the sequestering of dietary elements, causing a consumers odour to chemically match the odour of its prey. Chapter 4 tested for diet-induced chemical crypsis in O. longirostris by using the olfactory preferences of coral-obligate crabs to determine the effect of coral diet on fish odour. Crabs strongly preferred the odour of filefish fed their normal coral host, suggesting coral-specific dietary elements that influence odour are sequestered. In behavioural trials, predatory cod were less attracted to filefish odour presented alongside the coral it had been feeding on, suggesting reduced detectability. This evidence suggests that a close coupling between diet and habitat can form an effective antipredator strategy. Many animals select a habitat from the range of those available, driven by factors such as food quality or shelter availability, and these preferences may have consequences for the distribution and abundance of populations. Chapter 5 examined how the distribution and abundance of O. longirostris relates to coral architecture and diversity. The main drivers of distribution and abundance among reefs were coral species richness and availability of branching coral. Feeding territories had a higher percentage of Acropora coral than surrounding habitat. In addition, feeding territories had a higher percentage of the structurally important branching coral, A. nobilis, and the preferred prey species, A. millepora. Pair-wise choice experiments in which both structural complexity and coral tissue quality were independently manipulated showed that habitat choice was primarily based on shelter characteristics. These results suggest species-diverse coral habitats, which provide structural complexity along with nutritionally important prey, may be essential for population persistence. While extinctions of marine species are infrequent, local extinctions are becoming common. Yet, the role of habitat degradation and resource specialisation in explaining local extinction is still poorly understood. On coral reefs, coral bleaching is an increasingly frequent cause of coral mortality that can result in dramatic changes to coral community composition. Chapter 6 documents the local extinction of O. longirostris following a mass bleaching event. Local extinction occurred on reefs that lost all colonies of the key prey species, A. millepora, even though overall coral cover remained high. In an experimental test, fish continued to select bleached A. millepora over non-preferred species, irrespective of their condition. These results suggest that behavioural inflexibility may limit the ability of specialists to cope with even subtle changes to resource availability. This study demonstrates that O. longirostris is a highly specialised species that has evolved precise behavioural mechanisms to exploit the corals with which it associates. Its reliance on a relatively limited suite of corals leaves it highly vulnerable to climate change and other anthropogenic disturbances that threaten to reduce coral cover and abundance. This study also highlights how variable corals can be with regards to their value as prey and habitat. It appears that habitats that contain a specific combination of corals may be necessary for the survival and persistence of O. longirostris. When assessing the vulnerability of such highly specialised species, it is important to understand and evaluate the full range of their critical resources