Cetacean population monitoring : the merits and challenges of combining acoustic and visual data

Overview & Synopsis

Small cetaceans are vulnerable to anthropogenic pressures and their absolute abundance is needed to manage conservation efforts. The line-transect distance sampling method provides a reliable abundance estimate by determining the detection probability of animals at different distances from the transects. Detections on the transect line are assumed to be certain, i.e. g(0) = 1, but this is generally not the case due to detection biases. To address this, passive acoustic and visual surveys are used to provide complementary data although they are rarely combined. This thesis aims to exploit the strengths of both methods to improve detectability and abundance estimates. The vessel of the ACCOBAMS Survey Initiative operated continuously using acoustic and visual platforms either together or separately. Detection probabilities for each platform were estimated using the double platform framework. The g(0) was estimated to be 0.52 for combined platforms, 0.33 for acoustic and 0.30 for visual platforms. A significant diel pattern in acoustic events was observed with higher number of events at night. This variability was incorporated into abundance estimates, resulting in an estimated 847,298 small cetaceans in the Western Mediterranean, compared to 518,920 without detection biases correction. This study showed that combining visual and acoustic data is mostly effective when detection biases inherent to each method are properly taken into account. Recommendations were made to improve the use of these platforms in cetacean population monitoring.