Introducing ecological uncertainty in risk sensitivity indices: the case of wind farm impact on birds

. In wind-farm impact assessments, it is useful to know the level of uncertainty that characterizes some key variables used to assess the sensitivity to risk in species of conservation concern.Here, we have introduced the use of the Shannon entropy as a value of ecological uncertainty of theprediction of the risk assessment index, obtained from two ecological traits (flight altitude and flighttype). We based our evaluation of risk sensitivity on sampling of three common raptor birds (Gypsfulvus, Falco tinnunculus, Buteo buteo) all co-occurring in a wind farm landscape (central Italy). Asto flight altitude, Gyps fulvus prefers the flight altitude category > 300 m, Falco tinnunculus categoriescloser to the ground, Buteo does not show clear preference for a particular flight altitude category. Asregards the flight type, Gyps fulvus showed significant preference for circular and thermal flight patterns, Falco tinnunculus for rapid horizontal/vertical flight patterns, while Buteo was found to preferconstant directions and circular and thermal flight patterns. Multiplication of the scores associatedwith these eco-behavioural traits by the number of recorded occurrences allowed estimation of risksensitivity used to compute the risk assessment index. We normalized the partial scores of the riskassociated with these traits with H' values, thus obtaining more reliable species-specific normalizedrisk indices. The greater the entropy, the greater the level of uncertainty associated with it, and thelower the reliability of the risk index. Therefore, the entropy associated with flight behaviour diversity(altitude or type) could be a proxy for assessing uncertainty in wind power impact assessment. Wethink that normalizing indices of risk by including a measure of uncertainty can support decisionmakers in bird conservation and wind farm management.