Wildlife response to human functional disturbance: a case study during the Anthropause

Animal species are subject to two main different forms of an thropogenic disturbance, i.e. structural (e.g., presence of bar riers or habitat fragmentation) or functional (e.g., actual human presence for hunting or recreational activities). With specific reference to the latter, animals adopt a series of behavioural adaptations to cope with human disturbance, for example by switching their activity patterns towards night-time to minimize the risk of being hunted, or to reduce the temporal overlap with human activities, especially in highly anthropized environments. Examining the degree of behavioural plasticity of animal spe cies in relation to modifications of anthropogenic disturbance is therefore essential to fully understand the impact of humans on animal behaviour, and the capability of the animal species to adapt to human pressure. In this context, the reduction in human activities during COVID-19 lockdowns, which has been particularly severe in some Countries as Italy, constitutes an unprecedented semi-experimental condition to examine the responsiveness of animals to a dramatic alteration of human pressure, namely the “Anthropause”. We investigated this aspect in an anthropized area of Italian Alps, with the specific aim to analyse the adaptive response of three mammal species, namely roe deer (Capreolus capreolus), red deer (Cervus elaphus) and European badger (Meles meles) to the variation in functional disturbance that occurred during the Anthropause. We performed our assessment in a peri-urban area of the Eastern Alps (Valle di Cembra, Trentino Alto-Adige), focusing on the patterns of visits by these three species to eight artificial foraging sites, which have been monitored by means of camera traps both during the lockdown (9 March–4 May 2020) and in the same period in 2019, the latter being used as a baseline for our empirical evaluation of animal responsiveness to alteration of human pressure. We have used an artificial intelligence (AI) algorithm (YOLOv4), combined with manual observation, to analyse over 130000 pictures collected in 2019 and 2020. We have analysed the circadian pattern of visits to the feeding sites by means of circular statistics (daily trend), while we have used Generalized Additive Mixed Models (GAMMs) to investigate the trend of use of feeding sites by the three target species over the period of analysis (i.e., 9 March–4 May 2020 compared with 9 March–4 May 2019). We found that all the three species increased their number of visits to feeding sites in 2020 compared to 2019. We did not detect any temporal circadian alteration in the patterns of visits from 2019 to 2020, with badgers and red deer remaining strictly nocturnal, while roe deer exhibited the typical peaks of visits at dawn and dusk, with the rest of activity concentrated in nightly hours. Instead, the trend of visits to feeding sites over the examined period changed both for roe deer and badger, but not for red deer. While the pattern of visits to these anthropogenic resources was mainly dictated by seasonality in 2019, in 2020 we observed a clear link with the intensity of the lockdown, i.e., an increase of visits over the strict lockdown period, followed by a reduction when some mining activities occurring in the immediate surroundings restarted. The absence of a clear change in the circadian pattern of visits, i.e., without an increase in daytime visits during lockdown — often time reported by several unusual observations of wildlife in that period — suggests that the adaptation towards noctur nality in these species is likely a fixed behavioural trait, which a relatively short alteration of human disturbance has not been able to release. The response obtained in badger and roe deer relative to the trend during the observation period denotes the high plasticity of these species, which have increased the use of a resource linked to anthropogenic risk in the absence of cues of anthropic functional disturbance (for example noises quarries or forestry works). Taken together, these results confirm what has already been observed in numerous other studies, namely that the response of wildlife to the so called Anthropause is articulated and complex, depending on the characteristics of the species and the time scale at which the analysis is performed.