Abstract

AbstractHarbor porpoises (Cetacea) are present in the North Sea throughout the year but periodically enter adjacent estuaries, which due to human activities are among the planet's most threatened aquatic systems. However, the occurrence of harbor porpoises in estuaries has rarely been studied. In this work, harbor porpoise occurrence at two stations in the anthropogenically modified Ems Estuary (Germany, Netherlands) was modeled using a machine learning approach (Random Forest) that drew on 8 years of acoustic monitoring data with C‐PODs together with environmental data. Harbor porpoises were present year‐round at both stations. According to the models, their detection was mainly explained by season, tide, and noise level, with the highest detection probabilities in spring, at high tide, and at low noise levels. The seasonal and tide‐dependent occurrence of harbor porpoises coincided with prey availability. Presumed feeding activity was detected in 47% of all harbor‐porpoise‐positive 10 min blocks and indicated the importance of the estuary as a regular feeding area. The elevated noise levels detected at one station were attributed to tidal‐induced currents and sediment movements. The results of this study can help to improve estuarine management through measures that include conducting dredging and disposal activities when harbor porpoise occurrence is less likely.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call