Establishing protocols to apply repellents while hazing crop pests: importance of habitat, flock size, and time on blackbird (Icteridae) responses to a drone capable of spraying

Context Drones can be used as frightening devices to resolve avian-agriculture conflicts. Blackbird (Icteridae) flocks respond to drones making them a suitable scare device to protect sunflower (Helianthus annuus), but with limited efficacy on large flocks. Integrating a non-lethal avian repellent on the same drone as used for hazing may increase efficacy, but responses of flocks towards drones with spraying capabilities need to be evaluated to inform application protocols. Aim We evaluated flock responses to a drone capable of spraying when first approached and with 10 min of hazing, to inform protocols for delivering repellents on agricultural landscapes. Methods We used eye-in-the-sky drones to video the drone with spraying capabilities and observed whether flocks took flight within 80 m (i.e. range of potential spray drift). We measured flight initiation distance (FID) when close approach occurred (i.e. drone ≤80 m from flock). While hazing, we piloted the drone to (1) repeatedly cut through a flock and create chaos or (2) move along the flock edge to herd birds out of target habitat (i.e. sunflower or cattail). We recorded abandonment, flock reduction, and return rate of birds in response to drone hazing. Key results The probability of a close approach was greater with birds in cattail than in sunflower, but habitat did not influence mean FID when the drone was within 80 m (FID = 40 m ± 14.3 s.d.), abandonment (31 of 60 flocks), or mean percentage flock reduction (50% ± 37 s.d.). FID was shorter with smaller flocks, later in the day, but abandonment increased with smaller flocks as the day progressed. Although 52% of flocks abandoned, 81% returned after the end of hazing. Flight path of the drone (i.e. chaotic or herding) did not affect abandonment or flock reduction. Conclusions Although blackbirds perceived the drone approach as riskier (>FID) while foraging in sunflower earlier in the day, increased abandonment that occurred later in the day was likely to be due to satiation and movement to night-time roosts, instead of hazing impacts. Birds in sunflower interspersed with cattail used local refugia until the threat passed, then resumed foraging. Implications If applying an avian repellent with a spraying drone, protocols should consider time of day, flock size, and habitat. When selecting a flight path, pilots need to be concerned only with optimizing spray drift to reach areas with foraging blackbirds.