Using two drones to simultaneously monitor visual and acoustic behaviour of gray whales (Eschrichtius robustus) in Baja California, Mexico

This study provides an initial demonstration of a combined two-UAV (Unmanned Aerial Vehicle) system for measuring the underwater source levels and behavioural context of vocal and non-vocal marine mammal signals, information that is highly ecologically-relevant in terms of understanding how a species interacts and copes with conspecifics and its acoustic environment. Although the calls of a few species are well known, major gaps exist in our knowledge about the relationship between vocal output and behavioural context, gender and age for most species. Accurate parameter estimates (e.g., typical source levels, frequency ranges, and temporal characteristics of animal sounds) relevant to their behaviour (activities such as foraging, migrating, mating, or parental care) are needed to establish use of critical habitats (when monitored by acoustics) or to assess potential effects of anthropogenic sound exposure (including reduction of the detection space of sounds used for communication). The emergence of UAVs provides new perspectives and data collection capabilities for marine mammal research. Although UAVs have been frequently exploited for visual observations of whales, most approaches for monitoring and recording sounds from individual whales are still performed using overside hydrophones from a boat or using acoustic tags attached to the animals. Laguna San Ignacio (LSI), Mexico, is one of the breeding and calving grounds of the Eastern North Pacific gray whale (Eschrichtius robustus) population. LSI area's restricted geography, combined with the short dive times of the whales, provided an excellent field laboratory to test a dual drone approach. We used two UAVs: one to obtain acoustic measurements close to the whales and another one to obtain overhead visual observations. For the acoustic drone, a hydrophone (SoundTrap) was suspended via a 2-m line to a waterproof UAV quadcopter (Swellpro), which has the ability to take off and land from the surface of water. Simultaneously, the visual drone (DJI) monitored the whales in the area. Between 27 February and 17 March 2019, we simultaneously recorded underwater gray whale sounds and visual behavioural observations. During 92 min of underwater acoustic recordings, the acoustic drone recorded 11 call types. By time-synching underwater audio with the behavioural video, we obtained new insights into the source levels and functions of various quiet underwater sound that are difficult to impossible to obtain with standard methods. To our knowledge, no studies combining overhead visual observations and underwater acoustic recordings to describe acoustic behaviour and sound parameters of calls have been previously published.