Advanced Technique for Automatic Detection and Discrimination of a Click Train With Short Interclick Intervals From the Clicks of Ganges River Dolphins (Platanista gangetica gangetica) Recorded by a Passive Acoustic Monitoring System Using Hydrophone Arrays

Abstract

AbstractTo understand the biosonar click characteristics of Ganges river dolphins (adults, young adults, and calves) in a wild environment along with periodic visual observations, an ongoing program for long-term in situ monitoring has been carried out using a passive acoustic monitoring (PAM) system. During monitoring phase 4 (2012), migrating Ganges river dolphin groups with small calves were visually observed, and click trains having a short interclick interval (ICI: from 6 to 12 ms) were concurrently found from the acoustic data corresponding to the period. Click trains having a short ICI have also been observed in other small-toothed whales during foraging and socializing activities (called buzz) (Thoms, Moss, & Vater, 2004; Simard & Mann, 2008). For analysis of the short ICI click trains produced by the Ganges river dolphins, an advanced technique that automatically detects and discriminates a “short ICI click train” from other click sequences during the selected periods of data recorded by a PAM system was developed. For a robust algorithm, a smaller mean ICI caused by overlapping click trains from multiple dolphins that fulfill the range of ICI values that may get incorrectly labeled as “short ICI click train” was considered to judiciously detect a reliable click train. By applying the selected parameters and ICI values (default or given), the performance of the proposed technique was demonstrated using sample data. The results showed the reliability of the technique for the extraction of a variety of short ICI click trains from other click trains.