Behavioural impact assessment of unmanned aerial vehicles on Weddell seals (Leptonychotes weddellii)

• Published in: Journal of experimental marine biology and ecology 2021-03, Vol.536, p.151509, Article 151509
• Main Authors: Laborie, Joris, Christiansen, Fredrik, Beedholm, Kristian, Madsen, Peter Teglberg, Heerah, Karine
• Format: Article
• Language: English
• Subjects: Antarctica; Behaviour; Drone disturbance; Marine mammals; Noise exposure; Pinnipeds
• ISSN: 0022-0981; 1879-1697
• DOI: 10.1016/j.jembe.2020.151509

The rapid increase in the use of unmanned aerial vehicles (UAVs) in wildlife research has raised concerns about its potential negative impact on animals. The paucity of studies and the variability of responses of pinnipeds to UAVs prompts the need for species-specific impact assessments. Here we assessed the potential behavioural impact of low altitude UAVs on Weddell seals (Leptonychotes weddellii). This is a preliminary step to envisage the feasibility of replacing and/or complementing traditional ground-based behavioural and morphometric measurements by potentially less invasive UAV aerial images. We flew a small UAV (DJI Mavic 2 zoom fitted with a phocid seal audiogram weighted source level of 84 dB re 20μPa (rms)) over 37 Weddell seals (3 adult males, 12 adult females and 22 mother-pup pairs) during the breeding season at Dumont D'Urville, East Antarctica. For each individual, we assessed the level of reaction during UAV overflights at three altitudes (25, 20 and 15 m) while factoring in pup presence and wind speed. For all altitudes and observations pooled together, Weddell seals predominantly (88%) showed little (vigilant) or no (resting) reactions towards the UAV. Moreover, only 27% of all individuals changed their initial activity during the sampling periods, and mothers rarely ended their nursing bouts (3%). While reactions were low overall, the probability of a stronger reaction occurring increased at lower altitudes, and varied among individuals. Neither the presence of pups nor a change in wind speed appear to influence individuals' response to the UAV significantly. However, on simpler histogram representations of the dataset, we observed the strongest reactions for females (n = 5) with a pup at wind speeds below 5 m.s−1 when ambient noise levels were lowest. While Weddell seals are likely to hear the UAV at 25 to 15 m altitude in low wind speeds, the low-level responses we observed are unlikely to negatively impact their energetic budget and/or reproductive success. Our results suggest a low impact of small UAV overflights of Weddell seals during the breeding season when flying ≥25 m. This allows for collection of high resolution images for behavioural and morphometric studies that can potentially replace more invasive data collection when capturing and handling the animals. •We investigated the effect of a small, low flying (15–25 m) drone on Weddell seals.•The drone had only a minor impact on the behaviour of the seals.•Lower drone flight