Predicting population-level risk effects of predation from the responses of individuals.

Fear of predation produces large effects on prey population dynamics through indirect risk effects that can cause even greater impacts than direct predation mortality. As yet, there is no general theoretical framework for predicting when and how these population risk effects will arise in specific prey populations, meaning that there is often little consideration given to the key role predator risk effects can play in understanding conservation and wildlife management challenges. Here, we propose that population predator risk effects can be predicted through an extension of individual risk trade-off theory and show for the first time that this is the case in a wild vertebrate system. Specifically, we demonstrate that the timing (in specific months of the year), occurrence (at low food availability), cause (reduction in individual energy reserves), and type (starvation mortality) of a population-level predator risk effect can be successfully predicted from individual responses using a widely applicable theoretical framework (individual-based risk trade-off theory). Our results suggest that individual-based risk trade-off frameworks could allow a wide range of population-level predator risk effects to be predicted from existing ecological theory, which would enable risk effects to be more routinely integrated into consideration of population processes and in applied situations such as conservation.

Reproductive Failure in UK Harbour Porpoises Phocoena phocoena: Legacy of Pollutant Exposure?

Reproductive failure in mammals due to exposure to polychlorinated biphenyls (PCBs) can occur either through endocrine disrupting effects or via immunosuppression and increased disease risk. To investigate further, full necropsies and determination of summed 25 polychlorinated biphenyls congeners (∑PCBs lipid weight) in blubber were undertaken on 329 UK-stranded female harbour porpoises (1990-2012). In sexually mature females, 25/127 (19.7%) showed direct evidence of reproductive failure (foetal death, aborting, dystocia or stillbirth). A further 21/127 (16.5%) had infections of the reproductive tract or tumours of reproductive tract tissues that could contribute to reproductive failure. Resting mature females (non-lactating or non-pregnant) had significantly higher mean ∑PCBs (18.5 mg/kg) than both lactating (7.5 mg/kg) and pregnant females (6 mg/kg), though not significantly different to sexually immature females (14.0 mg/kg). Using multinomial logistic regression models ΣPCBs was found to be a significant predictor of mature female reproductive status, adjusting for the effects of confounding variables. Resting females were more likely to have a higher PCB burden. Health status (proxied by "trauma" or "infectious disease" causes of death) was also a significant predictor, with lactating females (i.e. who successfully reproduced) more likely to be in good health status compared to other individuals. Based on contaminant profiles (>11 mg/kg lipid), at least 29/60 (48%) of resting females had not offloaded their pollutant burden via gestation and primarily lactation. Where data were available, these non-offloading females were previously gravid, which suggests foetal or newborn mortality. Furthermore, a lower pregnancy rate of 50% was estimated for "healthy" females that died of traumatic causes of death, compared to other populations. Whether or not PCBs are part of an underlying mechanism, we used individual PCB burdens to show further evidence of reproductive failure in the North-east Atlantic harbour porpoise population, results that should inform conservation management.