Coping with change in predation risk across space and time through complementary behavioral responses.

BACKGROUND: Our picture of behavioral management of risk by prey remains fragmentary. This partly stems from a lack of studies jointly analyzing different behavioral responses developed by prey, such as habitat use and fine-scale behavior, although they are expected to complement each other. We took advantage of a simple system on the Kerguelen archipelago, made of a prey species, European rabbit Oryctolagus cuniculus, a predator, feral cat Felis catus, and a mosaic of closed and open foraging patches, allowing reliable assessment of spatio-temporal change in predation risk. We investigated the way such a change triggered individual prey decisions on where, when and how to perform routine activities. RESULTS: Rabbit presence and behavior were recorded both day and night in patches with similar foraging characteristics, but contrasted in terms of openness. Cats, individually recognizable, were more active at night and in closed patches, in line with their expected higher hunting success in those conditions. Accordingly, rabbits avoided using closed patches at night and increased their vigilance if they did. Both day and night, rabbits increased their use of closed patches as compared to open patches in windy conditions, thereby probably reducing the thermoregulatory costs expected under such harsh environmental conditions. CONCLUSIONS: Overall, our data map the landscape of fear in this study system and indicate that prey habitat use and vigilance complement each other. Solely focusing on one or the other tactic may lead to erroneous conclusions regarding the way predation risk triggers prey decisions. Finally, future studies should investigate inter-individual variability in the relative use of these different types of complementary behavioral responses to perceived risk, along with the determinants and outcomes of such tactics.

From museums to genomics: old herbarium specimens shed light on a C3 to C4 transition.

Collections of specimens held by natural history museums are invaluable material for biodiversity inventory and evolutionary studies, with specimens accumulated over 300 years readily available for sampling. Unfortunately, most museum specimens yield low-quality DNA. Recent advances in sequencing technologies, so called next-generation sequencing, are revolutionizing phylogenetic investigations at a deep level. Here, the Illumina technology (HiSeq) was used on herbarium specimens of Sartidia (subfamily Aristidoideae, Poaceae), a small African-Malagasy grass lineage (six species) characteristic of wooded savannas, which is the C3 sister group of Stipagrostis, an important C4 genus from Africa and SW Asia. Complete chloroplast and nuclear ribosomal sequences were assembled for two Sartidia species, one of which (S. perrieri) is only known from a single specimen collected in Madagascar 100 years ago. Partial sequences of a few single-copy genes encoding phosphoenolpyruvate carboxylases (ppc) and malic enzymes (nadpme) were also assembled. Based on these data, the phylogenetic position of Malagasy Sartidia in the subfamily Aristidoideae was investigated and the biogeographical history of this genus was analysed with full species sampling. The evolutionary history of two genes for C4 photosynthesis (ppc-aL1b and nadpme-IV) in the group was also investigated. The gene encoding the C4 phosphoenolpyruvate caroxylase of Stipagrostis is absent from S. dewinteri suggesting that it is not essential in C3 members of the group, which might have favoured its recruitment into a new metabolic pathway. Altogether, the inclusion of historical museum specimens in phylogenomic analyses of biodiversity opens new avenues for evolutionary studies.

Analyzing the proximity to cover in a landscape of fear: a new approach applied to fine-scale habitat use by rabbits facing feral cat predation on Kerguelen archipelago.

Although proximity to cover has been routinely considered as an explanatory variable in studies investigating prey behavioral adjustments to predation pressure, the way it shapes risk perception still remains equivocal. This paradox arises from both the ambivalent nature of cover as potentially both obstructive and protective, making its impact on risk perception complex and context-dependent, and from the choice of the proxy used to measure proximity to cover in the field, which leads to an incomplete picture of the landscape of fear experienced by the prey. Here, we study a simple predator-prey-habitat system, i.e., rabbits Oryctolagus cuniculus facing feral cat Felis catus predation on Kerguelen archipelago. We assess how cover shapes risk perception in prey and develop an easily implementable field method to improve the estimation of proximity to cover. In contrast to protocols considering the "distance to nearest cover", we focus on the overall "area to cover". We show that fine-scale habitat use by rabbits is clearly related to our measure, in accordance with our hypothesis of higher risk in patches with smaller area to cover in this predator-prey-habitat system. In contrast, classical measures of proximity to cover are not retained in the best predictive models of habitat use. The use of this new approach, together with a more in-depth consideration of contrasting properties of cover, could help to better understand the role of this complex yet decisive parameter for predator-prey ecology.

Dealing with noisy absences to optimize species distribution models: an iterative ensemble modelling approach.

Species distribution models (SDMs) are widespread in ecology and conservation biology, but their accuracy can be lowered by non-environmental (noisy) absences that are common in species occurrence data. Here we propose an iterative ensemble modelling (IEM) method to deal with noisy absences and hence improve the predictive reliability of ensemble modelling of species distributions. In the IEM approach, outputs of a classical ensemble model (EM) were used to update the raw occurrence data. The revised data was then used as input for a new EM run. This process was iterated until the predictions stabilized. The outputs of the iterative method were compared to those of the classical EM using virtual species. The IEM process tended to converge rapidly. It increased the consensus between predictions provided by the different methods as well as between those provided by different learning data sets. Comparing IEM and EM showed that for high levels of non-environmental absences, iterations significantly increased prediction reliability measured by the Kappa and TSS indices, as well as the percentage of well-predicted sites. Compared to EM, IEM also reduced biases in estimates of species prevalence. Compared to the classical EM method, IEM improves the reliability of species predictions. It particularly deals with noisy absences that are replaced in the data matrices by simulated presences during the iterative modelling process. IEM thus constitutes a promising way to increase the accuracy of EM predictions of difficult-to-detect species, as well as of species that are not in equilibrium with their environment.