The Australian Drinking Water Guidelines (ADWG) Framework for the Management of Drinking Water Quality (NHMRC, NRMMC 2011) promotes a structured and systematic preventative management approach to assure safety of drinking water at point of supply. Central to the ADWG Framework are 12 Elements considered good practice by which to manage drinking water supply from catchment to tap. Application of key ADWG Elements to the treatment and distribution components of a water supply system tends to be relatively straightforward, but less so to the source water catchments and raw water reservoirs components. This can lead to a breakdown in the integrated approach for managing water contamination risks across all stages of water supply. Accordingly, the leading aim of this study was to practically demonstrate how Element 2 (Assessment of the drinking water supply system) of the ADWG framework can be applied to a source water catchment to improve recognition and knowledge of water quality at this key juncture in the water supply system.

Adoption of camera trapping as a survey method by wildlife practitioners is increasing at warp speed. The technique is now widely cited in the published scientific literature and it has quickly become an important and widely used method in wildlife research, wildlife monitoring, and citizen science. Camera traps have largely been developed as a tool satisfying the demands of a very large hunting industry in North America. Until recently, the needs of ecologists and wildlife enthusiasts had been second to those in pursuit of hunting trophies, and as such many camera trap models failed the litmus test for fauna surveillance. The magnitude of these limitations has not been adequately recognised by practitioners and has led to the adoption of the technique without full understanding of the constraints of the sampling tool. In this dissertation I aimed to highlight and resolve some of the pitfalls that practitioners face when sampling wildlife using camera traps. I provide a historical context summarising how methods have developed over the last decade and tried to redress some of the ongoing problems identified in the camera trap literature. To this end I provide advice and guidelines to help camera trap practitioners design studies, implement sampling and reporting on their findings. However, the main focus of my research has been to address the differences between camera trap models and brands, the biases of the equipment, the effects of placement and orientation on detection, the challenges of identification and species in photographs, and have instigated the development of computer assisted technologies that will revolutionise how wildlife researchers analyse camera trap image data. I have also used my research to provide constructive design advice to camera trap manufacturers to encourage better designs to suit the needs of wildlife practitioners. Recommendations are provided on what practitioners would consider the features of an ultimate camera trap design that have led to the development of two new models of camera traps, and modifications to existing models.

This is the first comprehensive camera trap study to examine hollow usage by wildlife in the canopy of trees. Eighty cameras directed at tree hollows were deployed across eight sites in nine species of eucalypt in north-east New South Wales. In total, 38 species (including 21 birds, 9 mammals and 8 reptiles) were recorded at hollow entrances over a three-month period. There was a significant difference between wildlife hollow usage associated with site disturbance and tree growth stage (ANOSIM, P>0.05); however, there was no significant difference associated with tree hollow diameter (ANOSIM, P>0.05). The level of anthropogenic disturbance at each site, including vegetation modification of the understorey, was a significant predictor of species presence. Despite the limitations of using camera traps in the canopy of trees this study demonstrates the potential to garner useful insights into the ecology and behaviour of arboreal wildlife.

Mammal mycophagy (consumption of fungi by mammals) is an important process in forested ecosystems around the world. Of great interest to ecologists are those mammals that excavate and consume the below-ground truffle-forming fungi that are symbiotic with forest trees. By dispersing ingested spores a vital ecosystem function is performed by these mammals. Despite this importance, virtually nothing is known about how quickly a truffle patch is discovered and depleted by mammals, how different mammal species share a common food resource, or how truffles are excavated and handled by mycophagous mammals. Using passive infrared (PIR) video camera traps, we studied truffle excavation by mammals in two widely separated temperate ecosystems: (1) Conifer Forest in New Brunswick, Canada; and (2) Eucalyptus Woodland in Tasmania, Australia. Our results show that mammals discover and deplete localised truffle resources rapidly, and that very different mammals in both ecosystems (squirrels and voles in Canada; potoroos in Australia) respond similarly to the presence of truffles in terms of foraging rates and activity patterns. The technique yielded a novel dataset on truffle excavation by mammals and the first quantitative data on visitation rates to truffle patches by a range of mammal species, throwing light on how mammals exploit this food resource.

Camera trapping in scientific research has captivated practitioners globally and is now widely used as a primary survey method despite the unknowns and uncertainties of the tools. Using photos to identify species, especially coexisting species of similar appearance and niche are fraught with danger and can lead to serious conservation and management outcomes if identification is incorrect. The aim of this investigation was to test how accurately mammalogists with expertise in wildlife surveys could identify a range of species that were recorded during camera trap surveys. The identification of small-to-medium sized Australian mammal species using camera trap imagery by 158 professional wildlife surveyors was investigated using an internet survey. Fifty eight questions were posed to assess practitioner expertise in mammal trapping, and their accuracy in identifying 21 photos of 10 small-medium sized mammal species. Particular focus was placed on the identification of the Hastings River Mouse ('Pseudomys oralis') but other rodent species such as the Black Rat ('Rattus rattus'), the Bush Rat ('Rattus fuscipes'), and the Swamp Rat ('Rattus lutreolus') were included. The survey indicated that the correct identification of small mammals is highly variable between images of the same species, and that as a whole the professional wildlife community performs poorly at the identification of such species. Identification was more accurate where species were less likely to be confused with similar looking species, or where their identification was simple and/or obvious.

Camera trapping is increasingly recognised as a survey tool akin to conventional small mammal survey methods such as Elliott trapping. While there are many cost and resource advantages of using camera traps, their adoption should not compromise scientific rigour. Rodents are a common element of most small mammal surveys. In 2010 we deployed camera traps to measure whether the endangered Hastings River mouse ('Pseudomys oralis') could be detected and identified with an acceptable level of precision by camera traps when similar-looking sympatric small mammals were present. A comparison of three camera trap models revealed that camera traps can detect a wide range of small mammals, although white flash colour photography was necessary to capture characteristic features of morphology. However, the accurate identification of some small mammals, including 'P. oralis', was problematic; we conclude therefore that camera traps alone are not appropriate for 'P. oralis' surveys, even though they might at times successfully detect them. We discuss the need for refinement of the methodology, further testing of camera trap technology, and the development of computer-assisted techniques to overcome problems associated with accurate species identification.

We undertook surveys of brush-tailed rock-wallabies ('Petrogale penicillata') at four colonies in Oxley Wild Rivers National Park, north-eastern New South Wales, with the aim of developing a technique based upon individual animal recognition that could be used to obtain robust population estimates for rock-wallaby colonies. We identified individuals on the basis of distinct morphological characters in each colony using visual observations, and used the data within a 'mark-recapture' (or sight-resight) framework to estimate population size. More than 37 h of observations were made over 10 sampling days between 18 May and 9 June 2010. We could identify 91.7% of all rock-wallabies that were independently sighted (143 of 156 sightings of 35 animals). A small percentage of animals could not be identified during a visit because they were seen only fleetingly, were in dense cover, or were partly obscured by rock. The number of new animals sighted and photographed declined sharply at the midpoint of the survey, and there was a corresponding increase in resighting of known individuals. Population estimates using the mark-recapture methodology were nearly identical to estimates of total animals seen, suggesting that this method was successful in obtaining a complete census of rock-wallabies in each colony.