Title

Improving estimates of absolute abundance of dugongs

Objective

To obtain estimates of the absolute abundance of dugongs for use in harvesting models to

estimate sustainable mortality levels.

Relevance

The existing aerial survey technique provides standardised indices of abundance only. Changes in population size can be detected using such indices, as it is not necessary toknow exactly how many animals are present as long as the size of the population is measured in a standardised way. We know that the present estimates are underestimates but due tothe unknown reliability of comparing surfacing behaviour as seen from land or a boat with that observed from an aircraft, we do not know by how much. Reliable estimates of absoluteabundance are required to estimate sustainable levels of mortality using established harvesting models.

Feasibility

The methodology outlined below was discussed at a workshop on dugong aerial survey techniques held at James Cook University in November 1997 and attended by statisticians withrelevant expertise, including Dr Ann Cowling (DMR CSIRO) and Professor Ken Pollock (University of North Carolina).

In preliminary research funded by Seaworld, a helicopter was used to obtain aerial video footage of dugongs in clear water in Moreton Bay to improve methods to obtain accurateestimates of dugong abundance. The video footage was used to study the surface and diving behaviour of individual dugongs as seen from an aircraft. Analysis of the video footageshows that this method can be used to document the amount of time dugongs spend on the surface of the sea in clear water. The disadvantage of this method is that the presence ofthe helicopter may induce dugongs to behave abnormally.

Dugong diving and surfacing behaviour can also be documented using time depth recorders (TDRs) on dugongs fitted with satellite transmitters, using established methodology. Thetemporal and depth resolution of the resultant data may be insufficient to estimate the amount of time dugongs spend on the surface. However, they should aid in the interpretationof the video footage.

Methodology

Focal dugongs (cows with attendant calves, calves, unaccompanied adults) will be videoed from helicopters in clear water at a range of depths, e.g. 0–5 m, 5–10 m, >10 mto estimate the probability of dugongs in various age/sex classes being sighted at the surface when all dugongs are potentially visible for various depth strata.

Time depth recorders

TDRs and satellite transmitters will be fitted to dugongs in various age/sex classes in an area where dugongs feed at a range of depths, such as Hervey Bay, to study the divingand surfacing behaviour at a range of depths.

The resultant data will be analysed in collaboration with Dr Ken Pollock, a statistician at the University of North Carolina to develop the statistical methodology required togenerate corrections for availability bias for the various depth/turbidity strata encountered during a survey.

Approximate cost

1. Obtaining and analysing video footage $60 000
2. Obtaining and analysing time-depth profiles using TDRs $80 000
3. Developing statistical models $20 000

Total $160 000

Time required

2–3 years