How do major roads affect Barn Owls? Distributin, space use, food source and mortality

Road network expansion is known as one of the main factors responsible for the decline of Barn Owl (Tyto alba) populations in Europe, although the full causes of this decline are still poorly understood. In this context we evaluated several issues related to Barn Owl’s ecology, in Southern Portugal, when interacting with major roads: (a) the effect of highway distance on owl’s occurrence pattern, (b) the behavioral in the vicinity of major roads, (c) the role of road verges as an attraction factor due to prey abundance, and (d) owls’ spatial and temporal patterns of road mortality. To assess species occurrence (presence/absence) we broadcasted Barn Owl adult calls in 122 sites at several distances from major roads and detected 47 individuals. From the set of environmental and road-related variables tested using logistic regression, only the distance to the highway showed a significant positive influence on Barn Owl presence (β= 0.995; p<0.05). Space use patterns and road crossing rates were investigated on the basis of seven radio-tagged individuals (4 ♂, 3 ♀), captured in their nests located close to the highway (<5 km). The crossing rate was low (one road crossing per 34.19 hours of radio-tracking). Of the tracked individuals only four (3 ♂, 1 ♀) had sample sizes strong enough to provide robust estimates of home-range size (Fixed Kernel Density Estimator 95%) and just a peripheral overlap between homeranges (ranging from 2.61 km2 to 9.37 km2) and the highway was observed. The overall abundance of small mammals, assessed through live-trapping, was significantly higher in highway verges (n=248) than in the two other dominant land uses (cork oak woodlands (n=35), and croplands (n=64), suggesting that road verges could be a suitable habitat for hunting. Between 2004 and 2007, 373 road-kills were detected in 314 km of national roads surveyed (0.30 Barn Owls kills.km-1.year-1). A higher number of casualties was registered in fall and winter months when the dispersion of juveniles occurs, while lower mortality frequencies were detected in the end of summer and early spring, corresponding to hatching and fledgling periods. Mortality hotspots revealed to be strongly related with altitude (β =-0.026; p<0.05), eucalyptus or pine forest cover (β =-0.001; p<0.05) and percentage of cropland areas crossed by the road (β =0.332; p<0.05). Although major roads do not seem to act as effective barriers to Barn Owls’ movements, their occurrence pattern is significantly affected by this linear structure. Moreover, the potential attraction effect due to higher prey density in the highway verges does not seem to have an effect in the foraging behavior of adults.