Zoological Society of London

Abstract An understanding of habitat use and levels of active behaviour is foundational to wildlife behaviour, ecology, conservation and management. These variables are commonly measured by tracking individuals in space and time using biologging. In principle, camera‐trap data also contain information about both activity level and habitat use; however, because motion‐triggered cameras capture only active animals, the raw rates of capture in different habitats give us an indication of habitat use by active animals only, missing patterns of use and preference by inactive individuals. Furthermore, capture rates are readily confounded by spatially imbalanced survey effort (either by design or by accident), and by habitat‐related variation in other determinants of trap rate (animal speed and camera sensitivity). Here, we show how camera‐traps can be used to infer patterns of both activity and habitat use by deconfounding these effects. Using simulation, we show that the method is minimally biased when the key underlying assumption (spatially invariant activity pattern) is met, but can otherwise be moderately biased. It also understandably fails to provide reliable information for parts of the day when there is very little or no activity, working best for populations that are at least partly available for detection throughout the day. We demonstrate the use of the method by applying it to 7 years of springtime data on seven mammal populations from habitat‐stratified camera‐trap monitoring in Hoge Veluwe National Park, the Netherlands. The method recovered credible patterns of habitat‐specific activity levels in all species, and credible patterns of habitat use and preference in all but one strictly nocturnal species. The method uncovered hidden diurnal migration patterns between habitats, preference for resource‐rich habitat, risk avoidance during public opening hours, and spatiotemporal segregation in some species. The new method allows analysis of habitat relationships for entire communities of larger mammals without the need for trapping and tagging.

Understanding the spatial and temporal distribution of hunting is essential for the management of protected areas, as well as for assessing its impacts on wildlife. This knowledge also allows patrols to be directed toward areas under greater hunting pressure. A promising approach to estimating hunting pressure is through the use of autonomous bioacoustic recorders combined with artificial intelligence tools. However, it is necessary to develop a process capable of accurately describing the spatial and temporal distribution of hunting and, in parallel with wildlife monitoring, understanding its ecological impacts. Therefore, our goal is to estimate hunting pressure across space and time and to assess its effects on forest ungulate species. We intend to evaluate the impact of hunting by testing its influence on the density, occupancy, and activity patterns of ungulates. To achieve this, we established a 1,500-ha sampling grid in Iguaçu National Park, where each of the 24 plots is equipped with one bioacoustic recorder and three camera traps, which will operate for two years. So far, we have completed four months of sampling. We also conducted a preliminary experiment to record gunshots from different types of firearms and ammunition at various distances from the recorders, aiming to model sound intensity as a function of distance. As a preliminary result, we found that gunshots from 12-gauge firearms can be detected with an intensity of 20 dB at distances of up to 900 meters, indicating this as an optimal spacing between recorders for triangulation of detected gunshots. uring the setup of the sampling grid, we already found seven clear pieces of evidence of poaching activity in different locations within the study area, including platforms used for poaching (2), campsites (2), artificial salt licks (1), and cartridges (2), and we recorded a poacher with a camera trap. In the coming months, hunting activity will be modeled spatially and temporally. The resulting database of bioacoustic recordings and camera-trap images will be made available to the scientific community to foster future collaborations and improve conservation strategies. Funded by FAPESP #2024/18433-5

Time-scaled phylogenetic trees summarising evolutionary relationships are fundamental to many analyses in biology, from diversification rate estimation to conservation prioritisation. The most comprehensive available summary of these relationships, the Open Tree of Life, synthesises information from over two thousand studies into a supertree covering the full range of global biodiversity, but its use in downstream analyses is limited by the lack of divergence times. Previous work has mapped dates from Open Tree's database of trees to certain nodes in the supertree, but for the majority of nodes no date is available. While algorithms exist to interpolate missing dates in a tree, we found that their time and memory requirements scaled quadratically with the number of nodes, which made it computationally infeasible to run them on the entire tree. In this work, we describe novel date interpolation algorithms that scale linearly with the number of nodes. These enabled us to produce a distribution of fully-dated trees containing 2.3 million extant described species, greatly expanding the scope of feasible phylogenetic analyses. We illustrate the utility of these trees by computing the most robust estimate yet of the phylogenetic diversity of the complete tree of life, incorporating both topological and temporal uncertainty.

Conserving tigers (Panthera tigris) in human-dominated landscapes requires careful management of human-tiger conflict. Tiger attacks on human are the extreme form of conflict occurring primarily during livelihood activities in the Terai arc landscape of Nepal. We assessed spatio-temporal patterns of tiger attacks on human and factors associated with human fatality due to attacks. We collected human tiger conflict data from the government authorities, primarily national park offices and division forest offices. We interviewed 245 victims or family members attacked by tigers between 2010 and 2024 using a semi-structured questionnaire. Most tiger attacks on human (>90%) occurred inside forest areas and over half were fatal. Men were attacked more frequently (65.3%) but fatality was higher among women (>70%). Majority of the victims (93%) were illiterate or had a low level of education and 62% were hill migrants while indigenous Terai peoples were less frequently (27%) attacked. Large number of the tiger attacks occurred during resource collection from forests or livestock herding. The victim’s group size, sex and posture during the tiger attacks affected human fatality. Tiger attack can be minimized by applying adaptive management of buffer zone and corridor forests, adopting safety protocols through behavior change campaigns, promoting alternative livelihoods and tiger tourism, and proactive management of conflict-causing tigers. The finding of this research has long term implications to strengthen human-tiger coexistence in human-dominated landscapes in Nepal and elsewhere. Keywords: Tiger attack, Human-tiger-conflict, Co-existence, Terai-arc-landscape, Nepal

Abstract The rapid decline of wild populations has resulted in complete extinction of many species. In some cases, this outcome has been avoided through the establishment of ex situ conservation breeding populations. These have the aim of safeguarding the survival of the species and ultimately their reintroduction to the wild. In the case of Partula tree snails, conservation work spanning 4 decades has included ex situ breeding and reintroduction attempts. Most Partula tree snails of French Polynesia are categorized as Extinct on the IUCN Red List as a result of the introduction of invasive predators. Ten species survive only in ex situ conservation breeding centres and reintroduction attempts have been underway since 2015. In 2024, a wild population of Partula tohiveana was located, formed as a result of reintroductions to Moorea Island over the previous decade. This is the first Extinct in the Wild invertebrate species to have been re-established through coordinated, interdisciplinary ex situ and in situ conservation actions.

Abstract Large image collections generated from camera traps offer valuable insights into species richness, occupancy, and activity patterns, significantly aiding biodiversity monitoring. However, the manual processing of these data sets is time‐consuming, hindering analytical processes. To address this, deep neural networks have been widely adopted to automate image labelling, but the impact of classification error on key ecological metrics remains unclear. Here, we analyze data from camera trap collections in an African savannah (82,300 labelled images, 47 species) and an Asian sub‐tropical dry forest (40,308 labelled images, 29 species) to compare ecological metrics derived from expert‐generated species identifications with those generated by deep‐learning classification models. We specifically assess the impact of deep‐ learning model architecture, the proportion of label noise in the training data, and the size of the training data set on three key ecological metrics: species richness, occupancy, and activity patterns. We found that predictions of species richness derived from deep neural networks closely match those calculated from expert labels and remained resilient to up to 10% noise in the training data set (mis‐labelled images) and a 50% reduction in the training data set size. We found that our choice of deep‐learning model architecture (ResNet vs. ConvNext‐T) or depth (ResNet18, 50, 101) did not impact predicted ecological metrics. In contrast, species‐specific metrics were more sensitive; less common and visually similar species were disproportionately affected by a reduction in deep neural network accuracy, with consequences for occupancy and diel activity pattern estimates. To ensure the reliability of their findings, practitioners should prioritize creating large, clean training sets and account for class imbalance across species over exploring numerous deep‐learning model architectures.

ABSTRACT Among the order Carnivora, there is variation in the extent to which species form social groups. Typically, such variation is often studied in the context of a single variable that may impact behavior. However, the multiple interacting processes that drive plastic behaviors such as sociality and space use are likely to operate over multiple temporal and spatial scales. To quantify the relative impacts of different drivers requires an integration of the different processes across the scales at which they operate. Here we use piecewise structural equation modeling (piecewise SEM) to quantify the direct and indirect impacts of variables on behaviors including social grouping, reproduction, and space use. We analyze the role of phylogeny, environment, and community structure on sociality. Home range, social group size, and diet were shown to be most strongly influenced by life history variables, whereas reproductive output was most strongly impacted by variation in the environment.