Ph.D. project by Edda Elísabet Magnúsdóttir
Long-term passive acoustic recordings using two data loggers
To continuously monitor the annual occurrence of cetaceans at the NE-coast of Iceland we deployed two seabed mounted Ecological Acoustic Recorders (EARs). The EAR is a microprocessor-based autonomous recorder. Each EAR is programmed to record sound up to 32 kHz for 1 minute every 15 minutes. Both EARs are retrieved every 5 months during a two-year period (Sept. ’08 – Sept. ’10). A custom MATLAB program is used to process the datasets automatically. The program extracts specific variables from each recording in order to determine a sound of interest. For abundance, the number of vocalizations in each time period (e.g., each day, each month, each ten-day period) is counted, providing a rough indication of the number of animals in the area (e.g., Širovi´c et al., 2008).
Visual surveys from land using a theodolite.
Visual observations from a land-based station were conducted from Oct. ’08 – Sept. ‘09. The cetaceans’ locations and movement were continuously mapped with a theodolite into a real-time mapping program (Cyclops-Tracker). Cyclops-Tracker automatically calculated the speed, diving time, course and re-orientation of individuals/groups. The data will be used to estimate habitat use, and matched with the EAR recordings to obtain species confirmation of the recorded sound. Since many species have different vocal behavior in the presence of different numbers of conspecifics (e.g., Parks et al. 2005) the number, location and behavior of sighted cetaceans will be compared to the sound type recorded.
Visual and acoustical line-transect surveys
The line-transect surveys will be conducted in Skjálfandi Bay to obtain more accurate estimates of cetacean abundance. Previous studies have shown that towed hydrophone arrays do detect more animals then a visual observer (Fristrup and Clark 1997; Mellinger and Barlow 2003). The line transect surveys will be performed once to twice a month from May – Sept. 2010. Standard visual survey methods will be used as described by Buckland et al. (2001). A hydrophone array will be towed behind the ship while performing the visual observations. Several different software programs, previously developed will be used for automatic detection of tooth whales. The PAMGUARD software will be used for post processing of raw wav file data and the MATLAB program for post processing of clicks trains to estimate animal location and perpendicular distance.