Assessing the short- and long-term response of gulls to non-lethal hazing on an offshore island

Non-lethal bird hazing techniques are commonly used to protect public health and safety, defend crops, and safeguard the birds themselves. Examples of their use include warding them away from airports, landfills, oil spills, and other avian toxic exposure situations. Herein we examine the efficacy of non-lethal hazing tools for minimizing impacts to Gulls Larus spp. prior to a proposed eradication of introduced House Mice Mus musculus at the South Farallon Islands, 30 km offshore of the Central California coast. Methods considered for removing mice include the aerial application of rodenticide, which poses an adverse risk to non-target wildlife, including Western Gulls L. occidentalis. During a 15-day hazing trial period conducted in late November and early December 2012, we evaluated the effectiveness of a combination of non-lethal wildlife hazing techniques, including biosonics, pyrotechnics, lasers, helicopter, and effigies for temporary reduction in gull attendance. We found that gull numbers continuously decreased during the trial period, achieving 92%-99.7% reduction in abundance during the last four days of hazing, relative to pre-trial counts. Gull attendance remained very low for at least four days after the cessation of hazing, and there was no evidence of habituation. We also compared counts from winter 2012 to counts during the same dates in 2010 and 2011 and concluded that our measures had reduced gull numbers by as much as 98% at the end of the hazing period. Variation in hazing efficacy was best explained by a model that included hazing method, cumulative day of trial, and time of day. Specifically, lasers, pyrotechnics, and techniques that combined auditory and visual stimuli had the greatest hazing efficacy. Our results demonstrate that non-lethal hazing can be highly effective at reducing gull numbers at roost sites in late November and December in central California, indicating substantial reduction in exposure risk during the proposed mouse eradication.