Abstract
The first offshore wind farm 'alpha ventus' in the German North Sea was constructed north east of Borkum Reef Ground approximately 45 km north off the German coast in 2008 and 2009 using percussive piling for the foundations of 12 wind turbines. Visual monitoring of harbour porpoises was conducted prior to as well as during construction and operation by means of 15 aerial line transect distance sampling surveys, from 2008 to 2010. Static acoustic monitoring (SAM) with echolocation click loggers at 12 positions was performed additionally from 2008 to 2011. SAM devices were deployed between 1 and 50 km from the centre of the wind farm. During aerial surveys, 18 600 km of transect lines were covered in two survey areas (10 934 and 11 824 km2) and 1392 harbour porpoise sightings were recorded. Lowest densities were documented during the construction period in 2009. The spatial distribution pattern recorded on two aerial surveys three weeks before and exactly during pile-driving points towards a strong avoidance response within 20 km distance of the noise source. Generalized additive modelling of SAM data showed a negative impact of pile-driving on relative porpoise detection rates at eight positions at distances less than 10.8 km. Increased detection rates were found at two positions at 25 and 50 km distance suggesting that porpoises were displaced towards these positions. A pile-driving related behavioural reaction could thus be detected using SAM at a much larger distance than a pure avoidance radius would suggest. The first waiting time (interval between porpoise detections of at least 10 min), after piling started, increased with longer piling durations. A gradient in avoidance, a gradual fading of the avoidance reaction with increasing distance from the piling site, is hence most probably a product of an incomplete displacement during shorter piling events.