Validating automated click detector dolphin detection rates and investigating factors affecting performance

Area of Science:

• Marine biology
• Bioacoustics
• Cetacean research

Background:

• Passive acoustic monitoring (PAM) is crucial for understanding cetacean distribution and habitat use.
• C-PODs are acoustic sensors used in PAM, but offshore detection rates and influencing factors require further study.
• Bottlenose dolphin (Tursiops truncatus) occurrence is of interest, particularly in relation to offshore developments.

Purpose of the Study:

• To empirically evaluate the offshore detection performance of C-PODs for bottlenose dolphins.
• To identify factors affecting C-POD detection probability in offshore environments.
• To validate C-POD reliability for long-term monitoring of dolphin presence.

Main Methods:

• Deployed C-PODs alongside archival acoustic recorders 12-30 km offshore in the Northwest Atlantic.
• Collected 6852 hours of acoustic data for comparison.
• Post-processed C-POD and acoustic recordings using PAMGUARD software.

Main Results:

• C-PODs demonstrated very low false positive rates (mean 0.003%) and high hourly detection accuracy (mean 99.6%).
• True positive detections were more likely with increased click detections and high-frequency noise (>20 kHz).
• C-PODs proved reliable for indicating dolphin presence at hourly or greater time scales.

Conclusions:

• C-PODs are highly accurate and reliable for detecting offshore bottlenose dolphins.
• Environmental acoustic factors, such as high-frequency noise, influence C-POD detection probability.
• C-PODs are suitable for PAM studies investigating dolphin occurrence patterns, including those near offshore windfarms.