With ever increasing demands to develop coastal environments for sustainable use, particularly within UK government’s renewable energy framework, policy makers and managers face growing pressure from conservation agencies as well as national and international regulatory bodies to comply with requirements to monitor the species and habitats in the areasunder development or allocated for future use.This work assesses one method of facilitating the monitoring of coastal cetacean species, and focuses onthe two most common cetacean species in Wales, bottlenose dolphin(Tursiops truncatus) and harbour porpoise (Phocoena phocoena), both listed in the Annex II of the EU Habitats Directive, which requires reporting on their favourable conservation status.The study focuses on one static acoustic monitoring (SAM) device, the C-POD, which is already widely used in cetacean monitoring studies due to its practicality, ease of use and inexpensive set-up. Despite its popularity, some questions on its capabilities remain unresolved. This work aims to answer some of those questions.
Firstly the thesis estimates the effective detection area for harbour porpoises with C-PODs and examines the performance of both the hydrophone’s click detectionas well as the automated train detection algorithm. This is achieved with two different field experiments, comprising of playing back both artificial and real harbour porpoise sounds to the C-PODs at increasing distances. The results illustrate the ability of the C-POD to effectively log clicks, and the comparative performance of the train classification software. The effectsof source level and distance from data loggeron the detectability of porpoisesare revealed,which enables the calculation of robustestimatesof effective detection areas for future acoustic monitoring studies of the species.Next the detection probability of bottlenose dolphins with C-PODs is examined using simultaneous visual observations and acoustic recordings and the effects of group size and behaviour on dolphin detection with acoustic devices is assessed. The analyses demonstrate a significant effect of both group size and behaviour on the detectability of dolphins consequently affecting the calculated effective detection radius and area. These findings pose a challengefordensity estimation of dolphins using an overall detection functionandemphasize the importance of prior knowledge of the activity and behavioural patterns of target animals for developing appropriate experimental design, effective placement of data loggers and a meaningful analysis of SAM data.The study also explored the difference inclick train characteristics between broadband hydrophone recordings and C-POD recordingsof bottlenose dolphins. Asignificant reductionin click rates was evident in the C-POD data, indicating that the C-POD train detection algorithm may not be logging the fullrange of dolphin echolocation, particularly those trains with long inter-click intervals.The thesis then goes on to compare C-POD data with visual observations to assess whether thedata can be used to discern behavioural information from the twospecies. The results confirm that regardless of the fact that the C-POD does not record the full vocal repertoire produced by these animals, it can reveal valuable behavioural information about its target species by detecting feeding or foraging events andidentifying important feeding areas, particularly for harbour porpoise but also potentially for bottlenose dolphin.
This thesis supports the use of C-PODs in future monitoring efforts of both bottlenose dolphins and harbour porpoises. The findings will help analyse and interpretexisting datasets and assist in designingeffective and useful monitoring surveys for harbour porpoises and bottlenose dolphins and other species with similar behaviour and vocalisationcharacteristics. Considering the already wide use of the deviceit is essential that future studies on the efficacy ofthe device continue. More detailed understanding of the effect of different behaviours and group sizes on the echolocation rate and characteristics for both specieswould be beneficial, particularly for the harbour porpoise. Further studies to extract dolphin feeding buzzes from C-POD data would increase the loggers’ ability to identify critical feeding sites. Improvements of the train classification algorithm would enhance the detection probability of dolphins with C-PODs, potentially overcoming the effects of behaviour and group size discovered here.