Quantifying dive behaviour and three-dimensional activity space of dugongs in the Gulf of Carpentaria

The dugong (Dugong dugon) is a listed species under the Commonwealth Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act). The Dugong has a high biodiversity value as the only extant marine herbivore and is culturally significant to many coastal indigenous groups in Australia (Marsh et al. 2012). They have sensitive life history traits which increases their vulnerability to population impacts. Aerial surveys have been the primary method to estimate distribution and abundance of dugong (Dugong dugon) in Australia. Abundance estimates from aerial surveys are significantly influenced by two main sources of bias that arise from; the probability of an animal being available to be detected (availability), and the probability of an animal being detected by the observer, given it is available for detection (perception) (Marsh & Sinclair 1989; Pollock et al. 2006). The bias associated with perception can be quantified and corrected for using a structured, methodological approach using multiple independent observers during data collection. However, quantifying the bias associated with availability (availability detection probability; ADP) requires additional data external to the survey, and requires information on dive behaviours of the target species.Estimating ADP in heterogeneous environments requires the quantification of two parameters; 1) the depth range at which a dugong is visible to aerial observers (i.e. detection zones), and 2) the proportion of time a dugong is likely to be present within the detection zone (Hagihara et al. 2014). Detection zones during aerial surveys are strongly influenced by environmental conditions, and therefore have the potential to significantly influence dugong population estimates. The Department of Environment and Natural Resources (DENR) manage a program that aims to improve dugong abundance estimates in the Northern Territory by accounting for dugong availability bias. To this end, time, depth and location data required to improve dugong availability estimates was collected using a dugong Secchi disk experiment and 10 telemetered dugongs within the Gulf of Carpentaria. This report provides the analyses andresults of these data to: 1) quantify the in-situ detectability of dugongs relative to regional bathymetry and time of day and, 2) Produce three-dimensional metrics and visualisations of telemetered dugongs