Description: We used our novel and programmable Porpoise Alarm (PAL, patd.) to synthesize life-like, electronic harbour porpoise communication signals based on those described for captive animals. In the Little Belt, Denmark, we employed PAL (source level 158 ± 1 dB p–p re 1 μPa@1 m; centroid frequency 133 ± 8.5 kHz) to synthesize three aggressive click train types termed “A”, “F3” and “M1” to naive, free-living harbour porpoises. Via theodolite tracking (372 h of total visual effort spread over 10 expeditions) we found that, depending on signal type, porpoises either avoid or become attracted to PAL: Signal types “A” and “F3” are slight deterrents, porpoises increasing minimum range (+23 to 32 m, respectively), whereas “M1” attracts porpoises, reducing range (by − 29 m). As determined via archival acoustic detectors (AADs), both signals “F3” and “M1” led the animals to significantly intensify their click rate (by +10% and 68%, respectively) while signal “A” led to a significant reduction ( − 59%). We propose that equipping fishing gear with PAL emitting signal “F3” could potentially reduce porpoise by-catch by increasing (1) awareness through enhanced echolocation and (2) distance to the nets. Detection probability and radius of PAL/AAD tandems could be improved by emitting signal “M1” to focus porpoise echolocation signals on the AAD. The signal may also be useful in luring animals away from hazards, which may be helpful for conservation measures prior to the onset of harmful acoustic activities such as pile-driving, seismic exploration or ammunition clearance.

Description: Every year, large numbers of harbour porpoises (Phocoena phocoena) perish accidentally in coastal and high sea commercial fishing gear as so-called bycatch. This bycatch can be significantly reduced through the deployment of pingers (acoustic deterrent devices). Due to the proven efficiency of pingers, they are now mandatory in several countries. However, concerns have been raised that the continuous source of noise might lead to habituation, habitat exclusion and noise pollution. It has not been resolved as to why harbour porpoises become entangled in gillnets. The porpoises might not pay sufficient attention to their surroundings or they do not echolocate continuously. Alerting the animals without deterring them seems, therefore, to be a good alternative. Negative effects of the pinger, such as habitat exclusion from important feeding grounds and underwater noise pollution would be reduced through this new method. Further, porpoises might have the chance to decrease the possibility of becoming entangled through an increased awareness to their surroundings. The aim of this thesis is to investigate the reaction of free-living harbour porpoises towards synthetic porpoise-like alerting sounds. The new PAL (Porpoise Alerting Device) simulating aggressive porpoise click trains (SL 154db ± 2dB p-p re lμPa at 1 m; 133 kHz ± 0,5kHz} was tested during six experimental trips from July - October 2012 in the Little and the Great Belt, Denmark. Predominantly, groups of two harbour porpoises (37.5 %; N=445) and single individuals (35.7 %) were sighted in the Little and the Great Belt during observations. Visual and acoustic data shows a decreasing trend of porpoise abundance after August. Significantly more harbour porpoise groups were sighted in the morning and in the evening compared to the mid-day period (p=0.02; N=4}. The working hypothesis is that the approaching harbour porpoises would increase their echolocation activity to identify the 'aggressor'. Three hypotheses were tested: 1) Active PAL will increase echolocation activity of harbour porpoises; 2) Active PAL will not increase harbour porpoise presence; 3} Active PAL will not attract harbour porpoises. Three different PAL signals (upsweep chirp, starting with 173 click/sec ending with 959 clicks/sec (resp. 437 to 774 clicks/sec), total 700 clicks; click train duration 1.2-1.3s; 6-7 repetitions/min; 15 min on, followed by 15 min pause} were tested. No significant difference in duration of porpoise presence and approach distance between PAL on and PAL off could be found, indicating the PAL signal does not deter the animals as opposed to traditional pingers. Unfortunately, harbour porpoises did not increase their echolocation activity while the PAL was active. Two PAL signals even led to a significant decrease in echolocation activity. The analysis of single events showed a variety of reactions, such as swimming direction towards the active PAL and shortly after away from the active device; increased echolocation activity right after the PAL started transmitting; increased echolocation activity right after PAL stopped transmitting. An increase in echolocation activity (p=0.048; N=24) after the PAL stopped transmitting led us to believe that an alteration of the signal to pause ratio, with only 3 signal repetitions per minute, could lead to an increase in echolocation activity. Recent results obtained in 2013 (Culik, pers. comm.) show a significant increase in echolocation activity while the PAL is active. Further research is needed to substantiate this result. Fishery trials will ultimately show if the PAL signal is able to significantly reduce porpoise bycatch.