Spatial ecology and conservation of the critically endangered swift parrot

Abstract

Conservation of highly mobile resource specialists depends on understanding where and when resources are available and how populations respond to resource configuration. These species are often resource specialists, which can make them vulnerable to resource bottlenecks in time and space. When they also have dynamic distributions, data collection and conservation planning is extremely challenging. Therefore, for species like the swift parrot, which is a highly mobile resource specialist with a dynamic distribution, ecologically relevant and spatiotemporally explicit estimates of distributions are urgently needed to guide conservation planning.

Prior to this research little was known of spatiotemporal variation in the distribution of the critically endangered migratory swift parrot in its breeding range. The swift parrot requires co-occurrence of two key functional habitats to breed (nesting and foraging) and relies on the flowering of Eucalyptus globulus and E. ovata for food. The overall aim of this research was to better understand and quantify the spatial ecology of the species to improve conservation planning and outcomes. The main impetus for this research was continuing extensive habitat loss (as a result of industrial-scale logging and land clearance) without an understanding of i) the importance of the loss of key sites or locations and ii) the implications of the discovery of novel predator during the course of the study.

Firstly, this thesis quantifies and describes a key functional habitat feature (i.e. nesting trees) to assist accurate identification of nesting habitat (Chapter 2). The research then uses data from a unique multi-year monitoring program to i) extend modelling approaches to account for imperfect detection and spatial autocorrelation, ii) quantify the strong link between changing food availability and the species distribution, and iii)quantify how this varies over time (Chapter 3). Then, using data sampled from each functional habitat the research quantifies annual change in the use, location and availability of functional habitats over the entire breeding range (Chapter 4). Finally, the abundance-occupancy relationship (AOR) is quantified temporally and spatially to better understand the implications of spatiotemporal changes in abundance and resource availability for the interpretation species distribution models (SDMs) (Chapter 5).

This research reveals highly aggregated nesting behaviour of the swift parrot at multiple spatial scales, and provides one of the first macroecological examples to quantify a direct link between the spatiotemporal distribution of a highly mobile species and food availability. This spatiotemporal variation in food not only means the availability of functional habitats can vary dramatically between years, but also that an increase or decrease in one functional habitat does necessarily correspond to a relative increase or decrease in the other. This has important ramifications for interpreting SDMs, identifying when and where resource bottlenecks may occur, and the assessment of exposure to other spatially variable threats (e.g. predation). Further, the research shows the AOR for mobile species in dynamic distributions can be highly variable over time and space. Importantly, the results also highlight that locations with high predicted occupancy and/or abundance do not necessarily equate to areas of high quality habitat. This thesis delivers some of the first fundamental and quantitative insights into the spatial ecology of highly mobile species that rely on variable environments, and provides guidance towards informing and developing conservation plans for this difficult to study group of species.