Assessing the potential contribution of roads to variation in British bird populations

Roads and their traffic are known to affect bird species at both individual and population levels. Collisions with vehicles can cause direct mortality, and noise, light and chemical pollution can have sub-lethal impacts. Additionally, roads can cause habitat degradation, fragmentation and edge effects. However, thus far, the published literature on this topic comprises only relatively small-scale studies and our understanding of the impacts of roads on bird populations at landscape or national scales is limited. I use bird count and road data from across Great Britain to assess the spatial associations between the density and traffic volumes of roads, and bird populations in the surrounding areas. In Chapter 1, I provide background detail on the impacts of roads on birds and introduce the premise and necessity of my thesis. In Chapter 2, I quantify changes in the detectability of birds in field surveys in relation to road exposure. I find that, while some species are significantly harder to detect in areas of higher road exposure, others are easier. I therefore suggest that, in analyses of bird populations around roads, where possible, variation in detectability with exposure to roads should be accounted for, to avoid under- or over-estimation of road impacts on birds. In Chapter 3, I incorporate my detectability models into a spatial analysis of bird populations and roads across Britain, for 51 common and widespread species. This methodology allows independent assessment of the associations between roads and bird abundance, accounting for the potentially confounding impacts of roads on detectability. I find the abundances of 30 species to vary significantly with exposure to roads, some negatively and others positively. Across the interquartile range of road exposure, the mean decrease in bird abundance (for species with significant negative associations) was -19% and the mean increase (for species with significant positive associations) was +47%. In Chapter 4, in order to explore interspecific variation in these associations, I analyse a further 24 rarer species, and then test my results for all 75 species in relation to five characteristics. In this analysis I find 58 species to vary significantly in abundance with exposure to roads and the mean changes in bird abundance across the interquartile range of road exposure to increase to -39% and +48%. I also find that, with increasing road exposure, species with higher national populations have relatively higher abundance, while nationally rarer species have relatively lower abundance. Smaller-bodied and migratory species are also more negatively associated with road exposure. The distances over which negative associations between road exposure and bird abundance can be detected reach a mean of 700 m from a road, an area covering over 70% of Britain and 41% of the total area of terrestrial protected sites. I suggest that roads may, like some other forms of human disturbance, create conditions that benefit generally common species at the expense of others, thus having the potential to contribute to large-scale simplification of bird communities. Finally, in Chapter 5, I give an overall discussion of my thesis and highlight the importance of further work to understand in more detail the impacts of roads on birds in Britain and elsewhere, and to mitigate them effectively.NER