Effects of human disturbance on detectability of non-breeding birds in urban green areas

Abstract

Animals adapted to disturbed habitats have evolved multiple behavioural strategies, spanning from hiding to displacing to less disturbed microhabitats. Urban areas pose new evolutionary challenges since animals often need to deal with novel environmental conditions. In this context, urban parks may constitute biodiversity hotspots within the concrete jungle. Nonetheless, the recent increase in recreational activities in urban parks potentially puzzles the ability of urban-dwelling animals to exploit these environments. In this study, we evaluated the effect of human disturbance and other contextual variables on the activity patterns of four bird species commonly found in European urban parks, covering a wide range of ecological characteristics: the blackbird (Turdus merula), the hooded crow (Corvus cornix), the Eurasian robin (Erithacus rubecula), and the wood pigeon (Columba palumbus). We performed repeated counts of these bird species in six urban parks in northern Italy and we fitted Bayesian N-mixture models to estimate the relationship between detection probability and human disturbance (number of people present in the park), phenology (date and time of the day), and weather conditions (temperature and precipitation). For all the species but the blackbird, we found a negative relationship between the number of people present in the park and the detection probability of the focal species. We also found species-specific effects of both phenology and weather conditions on the detection probability. Our results suggest that urban dwelling species can finely modulate their activity patterns in response to the level of human disturbance, suggesting a possible key role of behavioural phenotypic plasticity. Furthermore, uncovering patterns of detectability of urban fauna can help in planning biodiversity monitoring and conservation, as it provides useful information to carry out surveys when the probability of detecting individuals is highest, optimising resource investments and reliability of biodiversity estimates.

Keywords

• Detection probability
• Urban dwellers
• Non-breeding season
• Phenology
• Weather conditions
• Urban parks