Surfacing characteristics and diving behaviour of blue whales in Sri Lankan waters
详细信息 查看全文
- 作者：Asha de Vos ; Fredrik Christiansen ; Robert G. Harcourt ; Charitha B. Pattiaratchi
- 关键词：Balaenoptera musculus ; Markov chains ; Monte Carlo ; Northern Indian Ocean ; Surfacing behaviour
- 刊名：Journal of Experimental Marine Biology and Ecology
- 出版年：November, 2013
- 年：2013
- 卷：449
- 期：Complete
- 页码：149-153
- 全文大小：401 K
文摘
Surfacing behaviour and dive characteristics were quantified from focal follows of individual blue whales between January-March 2012 and 2013. During this period individual whales were followed from small boats to observe their surfacing patterns and breathing behaviour. Data on time at first surface, length of surface interval, number of blows, final dive time and whether or not the whale 鈥榝luked up鈥?before a deep dive were recorded. A step-wise modelling approach was used to estimate a number of surfacing characteristics: mean Inter-Breath Interval (IBI), bout duration and the number of surfacings in a bout. First, dives were classified as either surface dives or deep dives based on the occurrence of arching or fluking behaviour at the surface prior to a deep dive. The mean IBI of surface dives was 17.6 s (SD = 26.14) and for deep dives, 640.3 s (SD = 214.38). To account for temporal dependence between dive types, a first-order Markov chain was used to estimate the transition probability between dive types. Time series of dive types were then simulated, using Monte Carlo methods, while accounting for heterogeneity in IBI of the different dive types. The mean IBI of blue whales in Sri Lanka, obtained from the Monte Carlo methods, was 84.7 s (SD = 11.17). The mean bout duration was 145 s (SD = 28.31), with the mean number of breaths per surface bout being 9.3 (SD = 1.43). Whales also lifted their tail flukes out of the water on 55% of terminal dives, which is considerably more frequent than elsewhere in the world. These results significantly advance our understanding of blue whales in Sri Lankan waters. More specifically, this information is essential for the calculation of precise abundance estimates as it informs the detection probability parameters for line transect surveys. In this way it will help formulate better management decisions related to the conservation of this population.