Interacting lethal and nonlethal human activities shape complex risk tolerance behaviors in a mountain herbivore.

Animals perceive human activities as risky and generally respond with fear-induced proactive behaviors to buffer the circadian patterns of lethal and nonlethal disturbances, such as diel migrations (DMs) between risky places during safe nighttime and safer places during risky daytime. However, such responses potentially incur costs through movement or reduced foraging time, so individuals should adjust their tolerance when human activities are harmless, through habituation. Yet this is a challenging cognitive task when lethal and nonlethal risks co-occur, forming complex landscapes of fear. The consequences of this human-induced complexity have, however, rarely been assessed. We studied the individual DM dynamics of chamois (Rupicapra rupicapra rupicapra), 89 GPS-tracked individual-years, from/to trails in the French Alps in areas with co-occurring lethal (hunting) and nonlethal (hiking and skiing) disturbances, with different intensities across seasons. We developed a conceptual framework relying on the risk-disturbance hypothesis and habituation to predict tolerance adjustments of chamois under various disturbance contexts and across contrasted seasonal periods. Based on spatial and statistical analyses combining periodograms and multinomial logistic models, we found that DM in relation to distance to a trail was a consistent response by chamois (~85% of individuals) to avoid human disturbance during daytime, especially during the hiking and hunting periods. Such behavior revealed a low tolerance of most chamois to human activities, although there was considerable interindividual heterogeneity in DM. Interestingly, there was an increased tolerance among the most disturbed diel migrants, potentially through habituation, with chamois performing shorter DMs in areas highly disturbed by hikers. Crucially, chamois that were most human-habituated during the hiking period remained more tolerant in the subsequent harvesting period, which could increase their risk of being harvested. In contrast, individuals less tolerant to hiking performed longer DMs when hunting risk increased, and compared to hiking, hunting exacerbated the threshold distance to trails triggering DMs. No carryover effect of hunting beyond the hunting period was observed. In conclusion, complex human-induced landscapes of fear with co-occurring disturbances by nature-based tourism and hunting may shape unexpected patterns of tolerance to human activities, whereby animal tolerance could become potentially deleterious for individual survival.

Integrating temporal refugia into landscapes of fear: prey exploit predator downtimes to forage in risky places.

The landscape of fear is an important driver of prey space use. However, prey can navigate the landscape of fear by exploiting temporal refuges from predation risk. We hypothesized that diel patterns of predator and prey movement and space use would be inversely correlated due to temporal constraints on predator habitat domain. Specifically, we evaluated habitat selection and activity of the vicuña and its only predator, the puma, during three diel periods: day, dawn/dusk, and night. Pumas selected the same habitats regardless of diel period-vegetated and rugged areas that feature stalking cover for pumas-but increased their activity levels during dawn/dusk and night when they benefit from reduced detection by prey. Vicuñas avoided areas selected by pumas and reduced activity at night, but selected vegetated areas and increased activity by day and dawn/dusk. Vicuña habitat selection and movement strategies appeared to reduce the risk of encountering pumas; movement rates of pumas and vicuñas were negatively correlated across the diel cycle, and habitat selection was negatively correlated during dawn/dusk and night. Our study shows that an ambush predator's temporal activity and space use patterns interact to create diel refugia and shape the antipredator behaviors of its prey. Importantly, it is likely the very nature of ambush predators' static habitat specificity that makes predator activity important to temporally varying perceptions of risk. Prey which depend on risky habitats for foraging appear to mitigate risk by feeding when they can more easily detect predators and when predators are least active.

Temporal plasticity in thermal-habitat selection of burbot Lota lota a diel-migrating winter-specialist.

In this study, animal-borne telemetry with temperature sensors was coupled with extensive habitat temperature monitoring in a dimictic reservoir, to test the following hypotheses: behavioural thermoregulation occurs throughout the year and temperature selection varies on a diel and seasonal basis, in a winter-specialist diel-migrating fish. Burbot Lota lota demonstrated nightly behavioural thermoregulation throughout the year, with a large seasonal shift between selection for very cold temperatures (<2° C) optimal for reproduction during the spawning period and selection for warmer temperatures (12-14° C) optimal for hunting and feeding during non-reproductive periods. During daylight hours, while L. lota avoided habitats warmer than optimal for reproduction and feeding during the spawning and non-reproductive periods, respectively, active selection was limited to selection for 4-6° C habitat during the prespawning period. Although behavioural thermoregulation explained the night-time migration, behavioural thermoregulation only partially explained daytime behaviour, indicating that diel migration is best explained by a combination of factors. Thus, thermal-habitat selection was a good predictor of night-time habitat occupancy in a diel-migrating species. Together, these results show that thermal-habitat selection by fishes may be important throughout the year and a more seasonally plastic behaviour than previously recognized.

Vertical ecology of the pelagic ocean: classical patterns and new perspectives.

Applications of acoustic and optical sensing and intensive, discrete-depth sampling, in concert with collaborative international research programmes, have substantially advanced knowledge of pelagic ecosystems in the 17 years since the 1996 Deepwater Fishes Symposium of the Fisheries Society of the British Isles. Although the epipelagic habitat is the best-known, and remote sensing and high-resolution modelling allow near-synoptic investigation of upper layer biophysical dynamics, ecological studies within the mesopelagic and deep-demersal habitats have begun to link lower and upper trophic level processes. Bathypelagic taxonomic inventories are far from complete, but recent projects (e.g. MAR-ECO and CMarZ, supported by the Census of Marine Life programme) have quantitatively strengthened distribution patterns previously described for fishes and have provided new perspectives. Synthesis of net and acoustic studies suggests that the biomass of deep-pelagic fishes may be two to three orders of magnitude greater than the total global commercial fisheries landings. Discrete-depth net sampling has revealed relatively high pelagic fish biomass below 1000 m in some regions, and that gelatinous zooplankton may be key energy vectors for deep-pelagic fish production. Lastly, perhaps, the most substantive paradigm shift is that vertical connectivity among fishes across classical depth zones is prevalent- suggesting that a whole-water column approach is warranted for deep ocean conservation and management.