<Poster Session>Classification of stroke pattern of dugongs in Dungonab Bay, Sudan by using an acoustic data logger

Abstract

We conducted a series of acoustic tagging of dugongs to observe individual behavior. A male dugong was equipped with an underwater recorder, a GPS logger, a VHF transmitter, and a time-scheduled releaser in June, 2012, in Dungonab Bay in Sudan. Underwater sound, location, temperature and depth of tagged dugongs were obtained. A male dugong spent 96 % of their time in shallow waters (< 4m). There were sudden deep dives (> 20m) with the deepest dive at 40 m. Sounds of the dugong's tail stroke was detected by taking energy sum between 0 - 500 Hz at 50-ms interval. Distribution of inter-stroke interval showed tri-modal distribution with each peak characterized by acoustic energy of the tail stroke. Hence the stroke pattern of the dugong was categorized into 1) Slow and strong strokes and 2) weak and fast strokes. The acoustic energy and intervals of the tail stroke can therefore be assumed as quantity of motion. The quantity of motion decreased during the night and increased before dawn and dusk. This study suggests that daily energy expenditure by a dugong can be estimated by using the acoustic data loggers.