Quantifying disturbance to bottlenose dolphin social networks with increasing tour vessel activity in Shark Bay, WA
Foroughirad, V., Singh, L., Bejder, L., Krzyszcyzk, E., Patterson, E., Wei, Y. and Mann, J. (2013) Quantifying disturbance to bottlenose dolphin social networks with increasing tour vessel activity in Shark Bay, WA. In: 20th Biennial Conference on the Biology of Marine Mammals, 9 - 13 December, Dunedin, New Zealand.
For highly social animals such as bottlenose dolphins, survival likely depends on the strength and stability of social bonds. Previous research demonstrated the impact of boat-based tourism and provisioning on Shark Bay dolphin behavior and ranging, but no studies have examined how social bonds and structure are affected. Using 2068 hours of focal follow data, we constructed ego networks for 31 adult female bottlenose dolphins (Tursiops cf. aduncus) in the impacted area over 4 consecutive time periods (pre-tour vessels [T0], 1 tour vessel [T1 & T4], 2 tour vessels [T3]) spanning 23 years. Results indicate that with increasing tour boat activity, the ego network size tended to decrease (LMM, t=-1.784, p=0.079), while bond strength and network density increased (strength: t=2.296, p=0.025; density: t=2.459, p=0.017). Seven of the subjects in the study visit a beach daily to be fed fish by tourists (managed by the Department of Environment and Conservation). With increasing boat tourism, these provisioned dolphins increased their homophily with each other (t=5.481, p< 0.001), suggesting greater segregation from the remaining population. Comparison networks were constructed using survey data (1500 sightings) for the highly trafficked “impact” area and an adjacent sparsely trafficked “control” area over each of the four time periods. We compared metrics between 75 adult females in each area and under each impact level. Lower values for eigenvector centrality were obtained (permutation-based ANOVA, p=0.008) with increasing exposure to tour vessels in the highly trafficked area. We further explored changes in network structure over time by comparing the expected variation in community structure to the actual variation using cluster results from a dynamic community detection algorithm. These results demonstrate that even low levels of tour vessel activity can have a significant effect on social network structure for bottlenose dolphins, especially for already at-risk individuals such as provisioned animals.
|Publication Type:||Conference Item|
|Murdoch Affiliation:||School of Veterinary and Life Sciences|
|Item Control Page|