Offshore wind farms are the main project on Taiwan's western coast. Since the underwater noise generated by piling poses a threat to marine mammals, the issue of the detrimental impact of noise on Sousa Chinensis has drawn considerable attention. To avoid behavioral disturbances and injury from pile driving noise, we propose the Underwater Noise Impact Region Alert System (UNIRAS) to estimate the acoustic field at any depth and distance from the piling sources. The system can be illustrated as four components: environment databases, an acoustic propagation model, source modeling and alert region (AR) prediction. Inputs are derived from the Taiwan Coastal Ocean Nowcast/Forecast System (TCONFS), which generates underwater sound speed profiles with temporal and spatial variation, along with geoacoustic and bathymetry databases that are imported as environmental inputs. Adiabatic mode theory is used to simulate the piling noise propagation in shallow water and the impulsive noise emanating from the source is evaluated via the finite element method. With the auditory threshold of cetacean being set as the criterion level, the system can demonstrate the modeling outputs and predict the noise impact region, and these results are useful for planning how to station the guard boats to prevent dolphins from entering the noise impact region.
Automated classification of urban locations for environmental noise impact assessment on the basis of road-traffic content
