Investigating the impacts of grazing and fire on the endangered Alpine She-oak Skink, Cyclodomorphus praealtus.

Abstract

Understanding drivers of species decline is key to successful conservation management. However, identifying stressors and quantifying population decline can be complex and resource-intensive. Tools such as genetic analyses and occupancy modelling can provide cost-effective measures of population change. In this thesis, I investigated potential indicators of population status in a cryptic Endangered reptile, the alpine she-oak skink (Cyclodomorphus praealtus), with a focus on two globally relevant disturbance processes--grazing by exotic species, and fire. Cyclodomorphus praealtus occurs in the Australian Alps, on isolated plateaux in Victoria and on Kosciuszko Plateau in New South Wales. When I commenced this project, the species was classified as Data Deficient in New South Wales due to insufficient information on its distribution and threats to inform management. To assess the relationships between C. praealtus, grazing and fire, I first needed to examine the distribution of C. praealtus, and the impacts of grazing mammals across its potential habitat in New South Wales. Here, exotic herbivore populations are rapidly increasing, and fire frequency and intensity is expected to increase under a warming climate. I found that grazing activity of exotic herbivores, including functionally novel taxa, was greater and more extensive than native herbivores, and was associated with a more open ground layer of vegetation and higher weed cover. Second, I investigated the genetic status of C. praealtus. I found high population structuring with little admixture. Four populations were apparent on Kosciuszko Plateau and three in Victoria. Populations generally had low genetic diversity, high inbreeding, low allelic richness and small effective population sizes. I found no evidence of local adaptation and my results indicated that the species has a poor capacity to respond to substantial reductions in abundance and habitat loss. Third, I modelled C. praealtus detectability in relation to survey conditions, and occupancy in relation to site conditions, including fire and grazing. I established 120 fixed monitoring transects across the species' potential habitat in New South Wales. With assistance, I conducted 2,045 transect surveys over four years, detecting C. praealtus on 265 occasions at 52 sites. Site occupancy was unrelated to elevation, vegetation height or fire, suggesting the species can occupy a range of subalpine grassland environments. Interestingly, site occupancy was positively related to grazing activity. Due to the limited timeframe and a lack of pre-disturbance data, I could not determine the cause of this result but suggest that it reflects a correlative rather than causal relationship. The final paper of my thesis investigated C. praealtus morphology and behaviour in relation to grazing and fire. Importantly, body condition in C. praealtus declined strongly with increased grazing activity, and there was an increased prevalence of tail autotomy in areas subject to grazing and fire. As body condition and tail autotomy can influence breeding success and survival in reptiles, my results suggest a negative impact of grazing and fire on C. praealtus. Further, I consider these results in the context of the measures I investigated in the previous papers. I conclude that C. praealtus populations are at risk of decline, particularly when exposed to elevated levels of grazing. My research highlights the value of assessing a range of measures to evaluate species' responses to stressors and to evaluate the risk of decline. Combining individual-level and population-level measures may be particularly valuable when stressors are slow-acting and species responses are nuanced.