Spatial ecology and predatory behaviour of the introduced sugar glider in swift parrot breeding habitat in Tasmania
Abstract
Recent research revealed that predation by the introduced sugar glider (Petaurus breviceps) resulted in the failure of the majority (>80 %) of nesting attempts by the critically endangered swift parrot (Lathamus discolor) on mainland Tasmania during six consecutive breeding seasons. A population viability analysis conducted in 2014 indicated that the remaining swift parrot population will likely decline by 95 % during the next three generations (~15 years) due to the severity of predation at the nest. This thesis examines this unusual predator and its relationship with the swift parrot in two separate but related studies. The first was a regional scale survey of sugar glider occurrence in an important swift parrot breeding region, the highly fragmented Southern Forests. The main objective was to determine their detectability and site occupancy and examine whether the latter is related to hollow-bearing habitat extent at different spatial scales. Southern boobook call playback was employed to collect sugar glider presence/absence data across 100 sites (4-5 visits per site) during February/March 2016. Sugar gliders were recorded at 79 sites and their detectability was 0.53. At the landscape scale there was a weak relationship between sugar glider occupancy and extent of mature, hollow-bearing forest within 500 m of a site. Glider detectability was negatively influenced by wind speed. This study indicates that at regional scales, occurrence of sugar gliders in swift parrot habitat may be very high, and that the habitat characteristics preferred by swift parrots (mature, hollow-bearing forest) may also be positively associated with sugar glider occurrence.
The second study was a nest predation experiment conducted from February to April 2016 across six key swift parrot breeding regions on mainland Tasmania. The key objective was to determine if fine-scale habitat attributes or a food reward within a nest hollow affect the likelihood of nest predation to improve our understanding of what factors influence glider predatory behaviour around swift parrot nest hollows. I used motion activated camera-traps at known swift parrot nest hollows to test whether experimental treatment (deployment of a quail egg n = 29) increased predation by gliders relative to a control group (n = 33). At each nest, data on eight habitat variables were recorded. Sugar gliders visited 52 % of nests during the 45-day study period. The frequency of sugar glider attendance at experimental nests was not significantly different between control and treatment groups, but was positively related to nearby abundance of tree hollows. There was less support found for models that accounted for other microhabitat characteristics of sugar gliders (e.g. the presence of potential food plant species). This study suggests that factors other than the presence of a food reward (e.g. a bird nest) may drive the frequency with which sugar gliders inspect tree hollows for potential prey. The findings indicate that nests that occur in areas where tree hollows are more abundant may be more likely to suffer high predation. My study provides valuable information for the conservation of swift parrots. First, I demonstrate that sugar gliders are widespread throughout the swift parrot breeding range both at landscape (mainland Tasmania) and regional scales (within an important swift parrot breeding region). Second, I confirm that tree hollows are an important habitat characteristic that may affect the distribution of sugar gliders. Third, I show that sugar gliders do not visit experimental nests at different rates depending on whether or not a food reward (i.e. eggs) is present, rather, local forest characteristics may better account for differences in predation risk among swift parrot nests. My thesis contributes important data on the spatial ecology of sugar gliders in Tasmania and confirms the efficacy of a simple survey technique for a serious threatening process.