Abstract
Increased interest in renewable energy and newfound technological advances have led to recent development of offshore wind energy in the United States; however, most of what is known about offshore wind farm construction impacts on the maritime environment has been from Europe. Limited research has been conducted on the potential impacts on wildlife in the coastal waters of the United States, specifically on marine mammals. Large baleen whales, such as fin whales, are of particular concern due to limited knowledge about their populations, hearing abilities, and responses to human activities. They are susceptible to acoustic disturbances from the increased underwater anthropogenic noise generated from ocean activities. This study explores the occurrence of fin whales in the vicinity of offshore wind farm areas off the coast of New Jersey to help understand the anthropogenic impacts on fin whale ecology.
Approximately nine months of acoustic data from 2008 were collected by three Marine Autonomous Recording Units (receivers) off the coast of New Jersey that were located in what is now wind energy lease areas. The sound files from the receivers were investigated for the signature 20 Hz pulses that fin whales produce. Machine learning using automated detectors in the program Raven Pro captured data from acoustic events, and visual confirmations of spectrograms were applied in the analysis to determine fin whale presence, in which the date, duration of pulse train, and quantity of pulse trains in each day were recorded. Detector performance was evaluated so as to calculate the precision and recall rates.
Shipping traffic noise could potentially interfere with the receiver’s detection range of fin whales. Cargo ships were chosen to be investigated, and Automatic Identification System (AIS) data in 2009 and 2019 were used for traffic comparison. Using ArcGIS Pro, three different buffers with different diameter distances—one-kilometer, three-kilometer, and five-kilometer— were created around the geocode locations where the receivers had been placed. The frequency of cargo ships that traveled through each buffer distance was determined using the Intersect tool in ArcGIS Pro.