Birds are a major pest in vineyards both in New Zealand and overseas. There is a need for new behavioural research on birds' foraging habits and feeding preferences in vineyards, as much of the literature to date is anecdotal. Research on cues to birds' feeding will provide a basis on which new deterrent and control strategies can be devised. Spatial-and temporal bird damage in a small vineyard block was mapped to find if damage was correlated with grape maturity and environmental factors. Vineyard and field observations of bird behaviour using video technology combined with preference experiments aimed to establish the relative roles of grape sugar concentration and colour in avian selection. Proximity of vineyards to bird roosts affects damage levels, regardless of differing maturity between locations. The rate of damage tends to increase exponentially once grape maturity has passed a threshold of 13 °Brix. Bunches positioned closest to the ground receive more damage if blackbirds or song thrushes are the predominant pests. Both sugar concentration and grape colour were found to affect birds' feeding preference, but the importance of the two factors varied between years. Black and green grape varieties were differentially preferred by blackbirds (Turdus merula) and song thrushes (Turdus philomelos) while silvereyes (Zosterops lateralis) appeared to have no strong colour preference. It was apparent that there were other, not assessed, grape factors that also affect selection. In small unprotected vineyards that are adjacent to bird roosts the entire grape crop can be taken by bird pests. Besides removing the roosts, which can be beneficial shelterbelts in regions exposed to high winds, growers currently may have no alternative other than to use exclusion netting to keep crops intact. The differential preferences between bird species for variety characteristics suggest that any new deterrents and other strategies to deflect birds from grape crops may need to be species-specific.

L’urbanisation grandissante est certainement à l’heure actuelle parmi les phénomènes à l’origine des changements les plus importants induits par l’homme sur notre planète, et soulève donc de nombreuses questions concernant les conséquences d’une telle perturbation de l’environnement sur la biodiversité. La vie urbaine est souvent caractérisée par des conditions jugées comme contraignantes (fragmentation, dégradation et perte d’habitats, pollutions chimiques, sonores et lumineuses accrues, etc.) pouvant avoir des effets néfastes sur les vertébrés. Cependant, les mécanismes par lesquels l’urbanisation affecte la faune sauvage sont encore très mal compris. Dans ce contexte, l’objectif central de ce doctorat est d’améliorer notre compréhension de l’influence de l’urbanisation sur les espèces de vertébrés évoluant dans les environnements urbains. Pour cela, nous avons exploré, d’une part, l’influence globale de l’urbanisation, et d’autre part, l’impact spécifique de certaines contraintes caractéristiques des milieux urbains (pollution par les métaux lourds et pollution sonore) sur les passereaux, en utilisant une approche fonctionnelle intégrant notamment des études éco-physiologiques et éco-toxicologiques. Ainsi, nous avons mis en évidence que les passereaux urbains ne sont pas contraints énergétiquement par leur environnement pendant leur vie adulte, mais qu’en revanche, les conditions fortement modifiées des milieux urbains pourraient contraindre les individus pendant leur développement. De plus, nous avons montré que l’urbanisation est effectivement associée à une contamination par les métaux lourds plus importante, et surtout que des niveaux plus élevés de ces contaminants sont associés à des niveaux de corticostérone plus élevés (degrés de stress plus importants). Enfin, l’étude expérimentale de l’impact de la pollution sonore sur des paramètres très souvent négligés jusque-là (développement, physiologie, comportement anti-prédateur), nous a permis de montrer que le bruit d’origine anthropique pouvait avoir des effets particulièrement complexes sur le développement des vertébrés avec de possibles conséquences à long terme. L’ensemble de ces travaux soulignent l’importance d’évaluer l’impact des perturbations d’origine anthropique sur le développement phénotypique des individus afin de mieux comprendre l’influence de l’environnement urbain sur les populations de vertébrés. / Consistent expanding urbanization is certainly among the most important human-induced environmental changes facing our planet today, and thus raises important questions regarding the consequences of such environmental disturbance on biodiversity. Urban life is often characterized by constraining environmental conditions (e.g., fragmentation, degradation and loss of habitats, increased noise, light, and chemical pollutions) that can have detrimental effects on wild vertebrates. However, the mechanisms through which urbanization affects wildlife are still poorly understood. In this context, the main objective of this doctoral research is to improve our understanding of the influence of urbanization on vertebrate species that are able to live in urban environments. To this end, we explored both the global influence of urbanization and the impact of specific factors associated with urban life (heavy metals and noise pollutions) on passerine birds, using a mechanistic approach including eco-physiological and eco-toxicological studies. We demonstrated that urban passerines are not energetically constrained by their environment during their adult life, but conversely, that the altered environmental conditions of urban environments could constrain individuals during their development. Furthermore, we showed that urbanization is indeed associated with an increased heavy metal contamination, and most importantly, that higher levels of these contaminants are related to higher corticosterone levels (i.e., elevated stress levels). Finally, using an experimental approach to study the impact of noise pollution on several parameters that have often been overlooked (development, physiology, anti-predator behaviour), we showed that anthropogenic noise can have particularly complex effects on vertebrate’s development, with possible life-long consequences for developing birds. Overall, our results emphasize the importance of investigating the impact of human-induced environmental changes on the phenotypic development of individuals to better understand the influence of urban environments on vertebrate populations.