Expert opinion as a tool for quantifying bird tolerance to human disturbance

Human disturbance can have several adverse effects on wildlife and therefore is increasingly seen as a threat. A common resolution of problems associated with encroaching human activities is to separate them from sensitive wildlife areas by protective buffer zones or set-back distances within which human activity is restricted. The most common method to establish such protective regimes is to record empirically the distance at which animals show signs of disturbance to human activity. However, a literature review for 26 bird species revealed that in only six of these species were there empirical measures of disturbance distances when breeding, but buffer zones had been recommended or designated in all species, often in several instances. This inferred prescription of buffer zones despite a severe knowledge gap. As a research stopgap, for the 26 species, we surveyed over 1000 expert opinions which generated estimates of alert distance (AD) and flight initiation distance (FID) in response to an approaching human during incubation and chick-rearing. Surveyed opinions on FID were not statistically significantly different to empirical measures of FID. Opinions on AD were much greater than predictions based on body mass derived from a previous study, but other evidence inferred a problem with predictions rather than opinions. The slope of the relationship between opinions on AD and FID conformed to the 'fixed-slope rule' (i.e. FID = 0.44*AD) for incubating birds, but was higher for chick-rearing birds. At both stages of the breeding cycle, however, FID was approximately half-AD, reflecting previous studies. The validation exercises therefore provided some encouragement that the expert opinion survey produced realistic results, but we recommend that their use should be temporary until more empirical measures of disturbance distances are gathered. We further recommend that existing monitoring schemes in which field surveyors routinely visit birds' nests should incorporate protocols to measure disturbance distances to amass such information rapidly and in quantity. (c) 2008 Elsevier Ltd. All rights reserved.