Hodgson, Amanda Jane (2004) Dugong behaviour and responses to human influences. PhD thesis, James Cook University.

Preview	PDF Download (370kB)
Preview	PDF Download (5MB)

Abstract

Knowledge of the behavioural ecology of a species is important for the development of conservation initiatives. With an understanding of how behaviour has evolved under given environmental and phylogenic constraints, it is possible to predict the response of a population to novel circumstances such as anthropogenic disturbance. Little is known about many aspects of the behaviour of dugongs (Dugong dugon). This species is difficult to observe as dugongs are benthic feeders, usually occurring in turbid waters. They tend to be wary of boats or divers and individuals cannot easily be distinguished. As dugongs occur in shallow, coastal waters, they are particularly vulnerable to human impacts; however, these impacts have not previously been quantified through direct observations of dugong behaviour. To overcome the difficulties in observing dugongs, I developed a blimp-cam, which allowed me to video dugongs from a blimp tethered to my research vessel. The use of the blimp-cam was facilitated by selecting Moreton Bay, Queensland, as my study site, where dugongs are readily located in clear, shallow waters. I used this technology to obtain baseline information about dugong behaviour, and investigated the function of the large herds persistently formed by dugongs in Moreton Bay. I then observed the response of dugongs to boats and pingers (acoustic alarms used on fishing nets to reduce marine mammal bycatch) to determine the risk of boat strikes and effects of disturbance from these two sound sources. At a height of 50m the blimp-cam provided an overhead view of dugongs at water depths of up to 4m, and distances of up to 200 m via a monitor on board the research vessel. Using a remote control, I could scan large herds of dugongs or continually observe individuals. Through focal follows of individual dugongs I developed an ethogram and a daily time budget for dugongs in Moreton Bay. Dugongs spent most of their time feeding (41%), travelling (32%), and surfacing (ascending to, and descending from, surface, 18%), and relatively little time resting (7%), socialising (6%) or rolling (1%). Environmental variables accounted for little of the variability in the proportion of time dugongs spent in each behavioural category. Time budgets did not differ significantly between single individuals and mothers with calves. However, mothers spent significantly more time feeding and surfacing, and less time travelling than their calves. Calves were observed suckling for mean bout lengths of 87 s. The mean submergence time for all individuals was 75 s, but was significantly shorter for calves (72 s) in comparison to their mothers (82 s). Submergence times were not affected by depth (< or ≥ 1.5 m), but were affected by behaviour. Dugongs spent 3.5% of the day resting at the surface of the water, during which time they are particularly vulnerable to boat strike. Mother-calf pairs appear most vulnerable to boat strike because they spend more time near the surface than single individuals. Calves are especially vulnerable as they rise or submerge by crossing onto their mother’s back during a quarter of their dives, and spent 13% of their time travelling and resting over their mothers’ back. I found that individual dugongs spent significantly more time feeding while in large, dense herds than when in smaller groups or scattered, suggesting that these herds are formed primarily to facilitate feeding. Dugongs did not seek large herds for resting, and calves were less likely to be surrounded by dugongs other than their mothers, than single individuals. These observations suggest that dugongs do not shelter in herds when most vulnerable to shark attack, and that herds are unlikely to have a predatory defence function. Herd structure was fluid, with individuals changing nearest neighbours after an average of 1 min, and showing no obvious preference for nearest neighbour type (single individual or mother-calf pair). Thus there was no evidence of a social function for these herds. My results support the theory that seagrass distribution and seasonality, sediment type, a lack of other disturbance of seagrass beds, and a yearround presence of dugongs on the Moreton Banks facilitate cultivation grazing. Observations of the response of dugongs to boats passing opportunistically provided information on the risk of boat strikes. Unlike controlled passes which were restricted to the below-planing speed limit of the study area, independent boats were often travelling above this limit. Only boats travelling above planing speed were observed passing directly over the top of dugongs. I hypothesise that the distance of the flight threshold for dugongs remains constant regardless of boat speed. Thus the speed of an approaching boat determines the time dugongs have to evade the boat, and speed is the main factor affecting the risk of boat strikes. Controlled experiments were conducted to determine the effects of boats on dugong behaviour. The behaviour of focal dugongs during a 4.5 min time sample was not significantly affected by whether there was a boat passing, the number of consecutive passes made (1 to 5 passes), or whether the pass was continuous or included a stop and restart during the pass. During the subsurface interval of the focal dugong that corresponded with the control boat’s closest approach time, the travel distance, travel direction and subsurface time were not correlated with the boat’s approach distance. However, during this subsurface interval dugongs were less likely to remain feeding if the boat passed within 50 m than if it passed at a greater distance. Mass movements of dugong feeding herds in response to boats were obvious but only lasted an average of 122 sec. These movements occurred in response to boats passing at all speeds, and at distances of less than 50 m to over 500 m. Relatively low levels of boat traffic in Moreton Bay in winter mean that a maximum 0.8 – 6% of feeding time may be interrupted by boats. However, if the number of boats registered in Queensland continues to increase at the current rate, the rate of disturbance is likely to increase. The response of dugongs to pingers was tested to determine whether these alarms may prevent dugongs from using important habitat areas. An array of two 10kHz ‘BASA’ pingers did not cause an observable response by dugongs. There was no significant difference in the rate of dugong movement away from the focal arena surrounding the pingers, orientation of the dugongs, or the presence or absence of feeding plumes, while the pingers were active compared to when inactive. The observed responses suggest that boat strikes are currently a bigger threat to dugongs than disturbance from boats or pingers, and support speed restrictions for boats in areas commonly used by dugongs. My results also reflect the need for detailed risk assessments to be conducted in areas where dugong habitat overlaps with areas of high boat traffic, and prior to future developments that will increase boat traffic. Further studies that build on the fundamental knowledge of dugong behaviour gained through this research will provide an understanding of human impacts in a wide range of habitats and aid in developing appropriate anthropogenic mortality targets for dugongs.

Item ID:	73
Item Type:	Thesis (PhD)
Keywords:	Behavioural ecology, Anthropogenic disturbance, Dugong dugon, Blimp-cam, Dugong behaviour
Date Deposited:	20 Jun 2007
FoR Codes:	06 BIOLOGICAL SCIENCES > 0602 Ecology > 060205 Marine and Estuarine Ecology (incl Marine Ichthyology) @ 50% 06 BIOLOGICAL SCIENCES > 0602 Ecology > 060201 Behavioural Ecology @ 50%
SEO Codes:	96 ENVIRONMENT > 9608 Flora, Fauna and Biodiversity > 960808 Marine Flora, Fauna and Biodiversity @ 100%
Downloads:	Total: 5039 Last 12 Months: 69
	More Statistics