Habitat Selection Studies Research Articles

Process-focussed, multi-grain resource selection functions

Like most aspects of ecology, the process of habitat or resource selection scales in space as well as time. However, scaling questions have generally focused on extent including size of study area and home ranges that dictate availability of resources. Grain of analysis (size of resource units used) is generally restricted to questions of methodology as opposed to functional ecology. Most often, grain is adopted as a point, unit, or patch that is common in size to all habitat resources used and available; however, in the process of habitat selection, it is feasible that individual animals may opt to select for different resources at different grains. For example, animals may use units of vegetation association at a finer grain when feeding or resting compared to when moving through habitat. Here we introduce and evaluate the ‘multi-grain resource selection function’, or MRSF. We generated MRSFs for a case study of GPS-collared white-tailed deer (Odocoileus virginianus; n=14) at Riding Mountain National Park, Manitoba, Canada. We created models across two seasons and extents and varied the radius around used and available points within which resource types were measured, and compared models to evaluate the relative importance of resource variables at different grains. We hypothesized that resource selection would vary with grain and that RSFs computed using multiple grains would be more predictive than models computed using a single grain as they better incorporate the space of influence on decision making in different habitat areas. We found that models of animals using grains of different sizes for different resource types were characterized by comparatively lower AIC scores. We conclude that scaling grain can and should be considered in models of resource selection, and that animals make decisions on resource selection at multiple grains. The MRSF, like analyses incorporating individual effects, density dependence, and functional responses, brings us closer to incorporating process, rather than only patterns, into the study of resource and habitat selection.

Winter foraging site fidelity of king penguins breeding at the Falkland Islands

Foraging site fidelity has profound consequences for individual fitness, population processes and the effectiveness of species conservation measures. Accordingly, quantifying site fidelity has become increasingly important in animal movement and habitat selection studies. To assess foraging site fidelity in king penguins (Aptenodytes patagonicus) breeding at the Falkland Islands (51.48°S, 57.83°W), we measured overlap in time spent in foraging areas (at a 0.1° × 0.1° grid resolution) between successive foraging trips and foraging route consistency during the creche period. In total, 30 complete foraging trips from seven king penguins were recorded between April and October 2010. King penguins predominantly foraged on the highly productive Patagonian slope, to the north of the Falkland Islands [median foraging trip distance 213 km (SD = 215 km) and duration 12.8 days (SD = 14.7 days)]. Overlap in time spent in an area on consecutive foraging trips ranged between 2 and 73 % (mean 27 %, SD = 22 %). Bearing during the outbound portion of foraging trips was typically highly repeatable for individual birds, but foraging trip duration and distance were not. Travel during the outbound phase of foraging trips was consistent with the direction of the northward-flowing Falkland Current that may act as a directional cue or facilitate rapid transit to foraging areas. Flexibility in foraging trip distances and durations may be a response to changes in resource availability and changes in the energetic requirements of adults and chicks over an extended breeding cycle.

A multi-scale distribution model for non-equilibrium populations suggests resource limitation in an endangered rodent.

Species distributions are known to be limited by biotic and abiotic factors at multiple temporal and spatial scales. Species distribution models, however, frequently assume a population at equilibrium in both time and space. Studies of habitat selection have repeatedly shown the difficulty of estimating resource selection if the scale or extent of analysis is incorrect. Here, we present a multi-step approach to estimate the realized and potential distribution of the endangered giant kangaroo rat. First, we estimate the potential distribution by modeling suitability at a range-wide scale using static bioclimatic variables. We then examine annual changes in extent at a population-level. We define “available” habitat based on the total suitable potential distribution at the range-wide scale. Then, within the available habitat, model changes in population extent driven by multiple measures of resource availability. By modeling distributions for a population with robust estimates of population extent through time, and ecologically relevant predictor variables, we improved the predictive ability of SDMs, as well as revealed an unanticipated relationship between population extent and precipitation at multiple scales. At a range-wide scale, the best model indicated the giant kangaroo rat was limited to areas that received little to no precipitation in the summer months. In contrast, the best model for shorter time scales showed a positive relation with resource abundance, driven by precipitation, in the current and previous year. These results suggest that the distribution of the giant kangaroo rat was limited to the wettest parts of the drier areas within the study region. This multi-step approach reinforces the differing relationship species may have with environmental variables at different scales, provides a novel method for defining “available” habitat in habitat selection studies, and suggests a way to create distribution models at spatial and temporal scales relevant to theoretical and applied ecologists.

Habitat prioritization across large landscapes, multiple seasons, and novel areas: An example using greater sage‐grouse in Wyoming

ABSTRACTAnimal habitat selection is an important and expansive area of research in ecology. In particular, the study of habitat selection is critical in habitat prioritization efforts for species of conservation concern. Landscape planning for species is happening at ever‐increasing extents because of the appreciation for the role of landscape‐scale patterns in species persistence coupled to improved datasets for species and habitats, and the expanding and intensifying footprint of human land uses on the landscape. We present a large‐scale collaborative effort to develop habitat selection models across large landscapes and multiple seasons for prioritizing habitat for a species of conservation concern. Greater sage‐grouse (Centrocercus urophasianus, hereafter sage‐grouse) occur in western semi‐arid landscapes in North America. Range‐wide population declines of this species have been documented, and it is currently considered as “warranted but precluded” from listing under the United States Endangered Species Act. Wyoming is predicted to remain a stronghold for sage‐grouse populations and contains approximately 37% of remaining birds. We compiled location data from 14 unique radiotelemetry studies (data collected 1994–2010) and habitat data from high‐quality, biologically relevant, geographic information system (GIS) layers across Wyoming. We developed habitat selection models for greater sage‐grouse across Wyoming for 3 distinct life stages: 1) nesting, 2) summer, and 3) winter. We developed patch and landscape models across 4 extents, producing statewide and regional (southwest, central, northeast) models for Wyoming. Habitat selection varied among regions and seasons, yet preferred habitat attributes generally matched the extensive literature on sage‐grouse seasonal habitat requirements. Across seasons and regions, birds preferred areas with greater percentage sagebrush cover and avoided paved roads, agriculture, and forested areas. Birds consistently preferred areas with higher precipitation in the summer and avoided rugged terrain in the winter. Selection for sagebrush cover varied regionally with stronger selection in the Northeast region, likely because of limited availability, whereas avoidance of paved roads was fairly consistent across regions. We chose resource selection function (RSF) thresholds for each model set (seasonal × regional combination) that delineated important seasonal habitats for sage‐grouse. Each model set showed good validation and discriminatory capabilities within study‐site boundaries. We applied the nesting‐season models to a novel area not included in model development. The percentage of independent nest locations that fell directly within identified important habitat was not overly impressive in the novel area (49%); however, including a 500‐m buffer around important habitat captured 98% of independent nest locations within the novel area. We also used leks and associated peak male counts as a proxy for nesting habitat outside of the study sites used to develop the models. A 1.5‐km buffer around the important nesting habitat boundaries included 77% of males counted at leks in Wyoming outside of the study sites. Data were not available to quantitatively test the performance of the summer and winter models outside our study sites. The collection of models presented here represents large‐scale resource‐management planning tools that are a significant advancement to previous tools in terms of spatial and temporal resolution. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.

Winter habitat selection by Marsh Tits Poecile palustris in a British woodland

Capsule Radio-tracking and remote sensing showed Marsh Tits selected for English Oak Quercus robur trees within large winter home-ranges.Aims To investigate winter habitat selection by Marsh Tits in a British wood, testing for preferences in tree species and woodland structure.Methods Thirteen Marsh Tits were radio-tracked during the winter, and home-ranges were derived. Lidar and hyperspectral data were used to compare the vegetation structure and tree species composition of entire home-ranges and the core areas of intensive use within. Instantaneous sampling observations provided further information for tree species utilization.Results The mean home-range was very large (39 ha, n = 13). There were no significant differences in mean tree height, canopy closure, understorey height, or shrub volume between full home-ranges and the core areas of use. Core areas contained a significantly greater proportion of English Oak relative to availability in the full home-ranges. Instantaneous sampling confirmed that English Oak was used significantly more than other trees.Conclusion Selection for English Oak during winter contrasted with previous studies of breeding territories, indicating that habitat usage varies seasonally and demonstrating the need for habitat selection studies throughout the year. Large home-ranges help to explain the Marsh Tit's sensitivity to habitat fragmentation.

Using vertebrate prey capture locations to identify cover type selection patterns of nocturnally foraging Burrowing Owls.

Studies of habitat selection often measure an animal's use of space via radiotelemetry or GPS-based technologies. Such data tend to be analyzed using a resource selection function, despite the fact that the actual resources acquired are typically not recorded. Without explicit proof of resource use, conclusions from RSF models are based on assumptions regarding an animal's behavior and the resources gained. Conservation initiatives are often based on space-use models, and could be detrimental to the target species if these assumptions are incorrect. We used GPS dataloggers and digital video recorders to determine precise locations where nocturnally foraging Burrowing Owls acquired food resources (vertebrate prey). We compared land cover type selection patterns using a presence-only resource selection function (RSF) to a model that incorporated prey capture locations (CRSF). We also compared net prey returns in each cover type to better measure reward relative to foraging effort. The RSF method did not reflect prey capture patterns and cover-type rankings from this model were quite different from models that used only locations where prey was known to have been obtained. Burrowing Owls successfully foraged across all cover types; however, return vs. effort models indicate that different cover types were of higher quality than those identified using resource selection functions. Conclusions about the type of resources acquired should not be made from RSF-style models without evidence that the actual resource of interest was acquired. Conservation efforts based on RSF models alone may be ineffective or detrimental to the target species if the limiting resource and where it is acquired are not properly identified.

Uniting statistical and individual-based approaches for animal movement modelling.

The dynamic nature of their internal states and the environment directly shape animals' spatial behaviours and give rise to emergent properties at broader scales in natural systems. However, integrating these dynamic features into habitat selection studies remains challenging, due to practically impossible field work to access internal states and the inability of current statistical models to produce dynamic outputs. To address these issues, we developed a robust method, which combines statistical and individual-based modelling. Using a statistical technique for forward modelling of the IBM has the advantage of being faster for parameterization than a pure inverse modelling technique and allows for robust selection of parameters. Using GPS locations from caribou monitored in Québec, caribou movements were modelled based on generative mechanisms accounting for dynamic variables at a low level of emergence. These variables were accessed by replicating real individuals' movements in parallel sub-models, and movement parameters were then empirically parameterized using Step Selection Functions. The final IBM model was validated using both k-fold cross-validation and emergent patterns validation and was tested for two different scenarios, with varying hardwood encroachment. Our results highlighted a functional response in habitat selection, which suggests that our method was able to capture the complexity of the natural system, and adequately provided projections on future possible states of the system in response to different management plans. This is especially relevant for testing the long-term impact of scenarios corresponding to environmental configurations that have yet to be observed in real systems.

LiDAR Remote Sensing of Forest Structure and GPS Telemetry Data Provide Insights on Winter Habitat Selection of European Roe Deer

The combination of GPS-Telemetry and resource selection functions is widely used to analyze animal habitat selection. Rapid large-scale assessment of vegetation structure allows bridging the requirements of habitat selection studies on grain size and extent, particularly in forest habitats. For roe deer, the cold period in winter forces individuals to optimize their trade off in searching for food and shelter. We analyzed the winter habitat selection of roe deer (Capreolus capreolus) in a montane forest landscape combining estimates of vegetation cover in three different height strata, derived from high resolution airborne Laser-scanning (LiDAR, Light detection and ranging), and activity data from GPS telemetry. Specifically, we tested the influence of temperature, snow height, and wind speed on site selection, differentiating between active and resting animals using mixed-effects conditional logistic regression models in a case-control design. Site selection was best explained by temperature deviations from hourly means, snow height, and activity status of the animals. Roe deer tended to use forests of high canopy cover more frequently with decreasing temperature, and when snow height exceeded 0.6 m. Active animals preferred lower canopy cover, but higher understory cover. Our approach demonstrates the potential of LiDAR measures for studying fine scale habitat selection in complex three-dimensional habitats, such as forests.

Providing insights on habitat connectivity for male brown bears: A combination of habitat suitability and landscape graph-based models

The combined study of movement patterns and habitat selection, behaviors that are affected by landscape structure, is of critical importance for effective conservation of wildlife. In the same context, maintaining connectivity is a key component for facilitating movement and ensuring species persistence. The brown bear is a wide-ranging predator that moves across large areas, crossing various habitats of different quality. Here, we provide a methodological framework that allowed us to explore the relative importance of habitat quality upon movement patterns and connectivity. First, we developed habitat-suitability models by using telemetry data of eight male brown bears (12,893 radio-locations) collected in Greece. Next, we combined habitat-suitability maps with real movement data to develop graph models that represent movement networks as inter-connected patches of different quality. Network analyses revealed that movement networks demonstrate some well-studied properties regarding their structure and robustness (i.e. scale-free and small-world), providing insights on the contribution of patches of different quality to connectivity. Less suitable patches were found to play a critical role for facilitating brown bear's movement and landscape connectivity. Our findings raise the importance of considering the entire habitat of brown bear in conservation planning rather than isolated patches of high quality, as it might have been revealed by simple habitat-suitability models. Our results highlight the need of assessing the importance of the intervening matrix for facilitating connectivity and movement specifically in the case of large and highly mobile species that do not perceive landscape as strictly dichotomous.

Application of random forest algorithm for studying habitat selection of colonial herons and egrets in human‐influenced landscapes

AbstractUnderstanding the mechanisms of habitat selection is fundamental to the construction of proper conservation and management plans for many avian species. Habitat changes caused by human beings increase the landscape complexity and thus the complexity of data available for explaining species distribution. New techniques that assume no linearity and capable to extrapolate the response variables across landscapes are needed for dealing with difficult relationships between habitat variables and distribution data. We used a random forest algorithm to study breeding‐site selection of herons and egrets in a human‐influenced landscape by analyzing land use around their colonies. We analyzed the importance of each land‐use variable for different scales and its relationship to the probability of colony presence. We found that there exist two main spatial scales on which herons and egrets select their colony sites: medium scale (4 km) and large scale (10–15 km). Colonies were attracted to areas with large amounts of evergreen forests at the medium scale, whereas avoidance of high‐density urban areas was important at the large scale. Previous studies used attractive factors, mainly foraging areas, to explain bird‐colony distributions, but our study is the first to show the major importance of repellent factors at large scales. We believe that the newest non‐linear methods, such as random forests, are needed when modelling complex variable interactions when organisms are distributed in complex landscapes. These methods could help to improve the conservation plans of those species threatened by the advance of highly human‐influenced landscapes.

Faster the better: a reliable technique to sample anopluran lice in large hosts.

Among Anoplura, the family Echinophthiriidae includes those species that infest mainly the pinnipeds. Working with large hosts implies methodological considerations as the time spent in the sampling, and the way in that the animal is restrained. Previous works on echinophthiriids combined a diverse array of analyses including field counts of lice and in vitro observations. To collect lice, the authors used forceps, and each louse was collected individually. This implied a long manipulation time, i.e., ≈60min and the need to physically and/or chemically immobilize the animal. The present work described and discussed for the first a sample technique that minimized the manipulation time and also avoiding the use of anesthesia. This methodology implied combing the host's pelage with a fine-tooth plastic comb, as used in the treatment of human pediculosis, and keeping the comb with the lice retained in a Ziploc® bag with ethanol. This technique was used successfully in studies of population dynamic, habitat selection, and transmission pattern, being a reliable methodology. Lice are collected entirely and are in a good condition to prepare them for mounting for studying under light or scanning electron microscopy. Moreover, the use of the plastic comb protects from damaging taxonomically important structures as spines being also recommended to reach taxonomic or morphological goals.

Temporal scales of density‐dependent habitat selection in a large grazing herbivore

Habitat selection is a density‐dependent process, but little is known regarding how this relationship may vary across different temporal scales. Over long time scales, grazing shapes the structure, diversity and functioning of terrestrial ecosystems, and grazing‐induced changes in forage production over time are likely to affect the level of density dependence in habitat selection. In this fully‐replicated, landscape‐scale experiment, we investigated how density‐dependent habitat selection by a large grazing herbivore, sheep Ovis aries, develops over the time scale of a decade. We also address an often‐neglected challenge in habitat selection studies; namely, whether there is variation in use within a particular habitat or vegetation type and why. We found clear evidence of density dependence in habitat selection, with a wider use of habitats at high density. Despite a change in the standing biomass of high‐productivity vegetation at high herbivore density over the years, with herb biomass declining and graminoid biomass increasing, there was no clear evidence that these grazing‐induced changes in habitat over the years were strong enough to affect the level of density‐dependent habitat selection. The difference in selection for high versus low‐productivity habitats remained similar, despite annual fluctuations in the strength of selection. We found strong variation in selection within each vegetation type, even when vegetation types were mapped at a fine‐resolution scale. Our study shows that despite the interactive effects of herbivores and habitats, they are not always sufficiently strong enough to affect the level of density‐dependent habitat selection.

Food Habits of the Leopard Cat (Prionailurus bengalensis euptilurus) in Korea

Prionailurus bengalensis is divided into 12 subspecies from India and Southeast Asia to Russia, which vary clearly in appearance (Nowell and Jackson 1996; Groves 1997). Among them, P. b. euptilurus is reported to inhabit Manchuria, the northeastern part of China, East Russia, Tsushima Islands of Japan, and Korea (Nowell and Jackson 1996). Prionailurus bengalensis is widely distributed but threatened by habitat loss, hunting, and commercial trade in Southeast Asia. Currently, P. bengalensis has been listed and protected under Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES 2011). The subspecies P. b. bengalensis has also been designated as endangered by the United States Department of the Interior (Nowak 1999). In an attempt to understand the status of the wild population, several researchers have focused on this species’ distribution, potential impacts on population dynamics, and feeding behavior (Rabinowitz 1990; Tatara and Doi 1994; Grassman 2000; Rajaratnam 2000; Austin 2002; Grassman et al. 2005; Rajaratnam et al. 2007; Izawa et al. 2009). Prionailurus b. euptilurus became endangered in Korea because of secondary poisoning through un limited use of poisonous rodenticides, such as Fratol (sodium monofluoroacetate), after the Korean War (Won 1967). The leopard cat has been protected in Korea since 1998 from the risk of extinction from habitat loss and traffic accidents. According to Choi (2007), the 103 leopard cats were killed by car accidents from 2004 to 2006. Studies have been conducted in Thailand to investigate the ecological and dietary habits of P. bengalensis (Rabinowitz 1990; Grassman 2000; Austin 2002; Grassman et al. 2005), and feeding strategy studies of P. bengalensis inhabiting Tsushima Islands have been carried out in Japan (Inoue 1972; Tatara and Doi 1994; Izawa et al. 2009). In addition, feeding habits and habitat selection studies have been conducted on Iriomote cats (P. b. iriomotensis) and leopard cats (P. b. borneoensis) (Schmidt et al. 2003; Rajaratnam et al. 2007). Despite that P. bengalensis is a globally threatened species, it has only recently been studied internationally. Since P. b. euptilurus was first referred to as a Chosun animal species by Won (1968), taxonomic studies based on the appearance of P. b. euptilurus have been reported in Korea (Oh 2002). The home ranges and habitat selection of P. b. euptilurus that live in and around artificial landscapes (e.g., roads and small-scale ecological parks) have also been reported (Woo 2010). Until recently, only fragmentary studies of P. b. euptilurus, have become available, with very few basic ecological studies, and baseline data are required for this species in Korea. The aim of this study is to report basic ecological data and information about the prey species of P. b. euptilurus using fecal samples.

Accounting for individual behavioural variation in studies of habitat selection

A caribou wearing an animal-borne video camera (a) and animal-borne video footage taken from systems deployed on mule deer (Odocoileus hemionus) in north-central Washington state, USA (b-d). When paired with tracking technology, animal-borne video can reveal detailed information about behaviour and environmental features at each location: (b) feeding, (c) vigilant in the open, (d) vigilant in cover, (e) resting in the open. Accordingly, animal-borne video systems should allow for analyses of habitat selection by individuals in particular behavioural states. In Focus: DeCesare, N.J., Hebblewhite, M., Bradley, M., Hervieux, D., Neufeld, L. & Musiani, M. (2014) Linking habitat selection and predation risk to spatial variation in survival. Journal of Animal Ecology, 83, 343-352. Resource selection is often assumed to confer enhanced fitness, but this assumption is rarely examined. In a study involving woodland caribou subject to grey wolf predation, DeCesare et al. (2014) show that while patterns of selection by caribou did correspond with a fitness proxy (survival probability), individuals did not avoid wolf predation risk to the extent that would minimize mortality. Here, we use the results of this paper as a springboard for discussing the choice of fitness proxies and the need to account for individual behavioural variation in studies of resource selection.

Weighed down by science: do collar-mounted devices affect domestic cat behaviour and movement?

Context Animals carrying tracking and logging devices are subject to a range of instrument effects that negatively affect survival, reproduction and behaviour. The common recommendation is that device weight should not exceed 5% body mass (BM) for terrestrial species; however, this rule-of-thumb has little empirical basis. Modelling indicates that devices weighing less than 3% BM may still have impacts. Several studies have used telemetry and data loggers on domestic cats (Felis catus) with instruments ranging in weight from 30 g to 125 g, but there has been no quantitative evaluation of instrument effects. In addition, inexpensive GPS tags such as iGotU are increasingly being used to track domestic cats, but often with little acknowledgement of habitat-related location error. Aims We evaluated the impact of wearing devices of different weights on domestic cat movements, and quantified location error across typical suburban habitats. Methods We recorded movements of cats wearing three different GPS collar weights for a week at a time: light, 30 g (<1% BM); medium, 80 g (~2% BM); and heavy, 130 g (~3% BM). Location error (LE) and fix success rates (FSR) were compared between backyard habitats and up- or downward orientation of the GPS tags on collars. Key results Home-range size and distance travelled from home were smaller when cats wore the heaviest collar. LE was lower and FSR higher for GPS tags with direct-line-of-sight to satellites (e.g. on lawns), but there was no difference between tags placed in dense vegetation (hedges) or more open vegetation (trees), or tags oriented up or down. Conclusions Collars carrying instruments on cats should be no more than 2% BM (medium-weight collar). LE was large relative to typical urban habitat size, indicating that misclassification of locations into habitats could easily occur in habitat-selection studies. Implications Some published accounts of cat home-range sizes may be underestimates, resulting in underestimates of the extent of impacts on prey species. Habitat-use studies should acknowledge the error associated with GPS tags and incorporate it into analyses using techniques such as Brownian Bridges.

Likelihood‐based photograph identification: Application with photographs of free‐ranging bison

ABSTRACTUsing photographs to identify individual animals and monitor populations is becoming more common. However, photographic identification methods where measurements of morphological traits (e.g., horn length) are compared have received little attention. We present an approach for aiding with the identification of individual animals from photographs. The approach incorporates measurement data, metadata from photographs, and multiple sources of error, and calculates a matching score between pairs of photographs using a likelihood‐based algorithm. We tested and identified the false‐rejection error rate using 91 photographs, representing 33 known free‐ranging bison (Bison bison), and 117 simulated data sets with varying numbers of unique individuals, morphological measurements, and photograph error. We then used the approach to estimate the adult population size of bison in Prince Albert National Park, Canada, in 2011. For bison, the false‐rejection rate of our approach was 0.055. Using a Huggins closed population model with misidentification, we estimated 103 (95% CI = 82–130) and 46 (95% CI = 37–58) adult female and male bison, respectively. After incorporating field‐based calf‐ and juvenile‐to‐female ratios, we estimated 202 (95% CI = 171.6–231.4) bison. We found this estimate to be plausible using 2 minimum‐count aerial surveys conducted in March 2011 and 2012 for comparison. With our approach, researchers and managers can build capture histories of individuals, which can be used for studies of population dynamics and habitat selection. This approach can incorporate any morphological measurements extracted from photographs (e.g., coat color), making it robust to a variety of species and study systems. © 2013 The Wildlife Society.

Habitat selection by large herbivores in a southern African savanna: the relative roles of bottom‐up and top‐down forces

Animals must often balance food availability and predation risk when selecting habitat. Here, we examined habitat preferences of large mammalian herbivores in a long‐term fire experiment in the Kruger National Park, South Africa to assess the role of bottom‐up (e.g., forage quantity/quality) and top‐down (e.g., predation risk) processes in driving herbivore distributions. We focused on experimental plots (∼10 ha on average) that have been burned in the winter (August) since 1954 at 1‐ and 3‐yr intervals, or left unburned (n = 3 per burn type). Herbivore distributions (during both day and night) and plant community structure were surveyed on each plot during the growing seasons (November–March) of 2005–2006, 2006–2007, and 2007–2008. Overall, we sighted 4,187 individuals representing twelve species of mammalian herbivores. Impala (Aepyceros melampus), zebra (Equus quagga), and wildebeest (Connochaetes taurinus) comprised 37%, 28%, and 18% of all individuals observed, respectively. Several species such as African buffalo (Syncerus caffer), wildebeest, and giraffe (Giraffa camelopardalis) exhibited a significant trade‐off between food acquisition and minimizing predation risk by foraging in areas with lower density of woody vegetation. We also observed significant day vs. night dynamics in herbivore habitat selection. For example, zebra utilized annual or triennial burns during the day depending on which years the plots were burned, but they avoided triennial burns with dense woody vegetation in favor of more open annual burns at night when predators such as lions (Panthera leo) are more active. Similarly, the smaller, mixed‐feeding impala appeared to use riskier habitats with more diverse forage options during the day (triennial burns and unburned plots) but used less risky habitats at night (annual and triennial burns). Diurnal vs. nocturnal patterns are often overlooked in studies of habitat selection but are necessary for understanding the factors that shape distribution. The variation we observed in herbivore distribution patterns during this three‐year period indicates that different species exhibit different trade‐offs with respect to food and predation risk. Factors such as body size, nutritional requirements, prey escape tactics, and recent fire history likely mediated these interspecific differences.

Linking habitat selection and predation risk to spatial variation in survival

1. A central assumption underlying the study of habitat selection is that selected habitats confer enhanced fitness. Unfortunately, this assumption is rarely tested, and in some systems, gradients of predation risk may more accurately characterize spatial variation in vital rates than gradients described by habitat selection studies.2. Here, we separately measured spatial patterns of both resource selection and predation risk and tested their relationships with a key demographic trait, adult female survival, for a threatened ungulate, woodland caribou (Rangifer tarandus caribou Gmelin). We also evaluated whether exposure to gradients in both predation risk and resource selection value was manifested temporally through instantaneous or seasonal effects on survival outcomes.3. We used Cox proportional hazards spatial survival modelling to assess the relative support for 5 selection- and risk-based definitions of habitat quality, as quantified by woodland caribou adult female survival. These hypotheses included scenarios in which selection ideally mirrored survival, risk entirely drove survival, non-ideal selection correlated with survival but with additive risk effects, an ecological trap with maladaptive selection and a non-spatial effect of annual variation in weather.4. Indeed, we found positive relationships between the predicted values of a resource selection function (RSF) and survival, yet subsequently incorporating an additional negative effect of predation risk greatly improved models further. This revealed a positive, but non-ideal relationship between selection and survival. Gradients in these covariates were also shown to affect individual survival probability at multiple temporal scales. Exposure to increased predation risk had a relatively instantaneous effect on survival outcomes, whereas variation in habitat suitability predicted by an RSF had both instantaneous and longer-term seasonal effects on survival.5. Predation risk was an additive source of hazard beyond that detected through selection alone, and woodland caribou selection thus was shown to be non-ideal. Furthermore, by combining spatial adult female survival models with herd-specific estimates of recruitment in matrix population models, we estimated a spatially explicit landscape of population growth predictions for this endangered species.

Return of the pythons: first formal records, with a special note on recovery of the Burmese python in the demilitarized Kinmen islands

We report the first formal records and population recovery of wild Burmese pythons (Python molurus bivittatus) in the Kinmen islands, which are located 2.1 km from the Chinese coastline and are under military control of the Taiwanese government. During the Cold War period, bombardments by China destroyed most of the buildings and vegetation on the island, possibly extirpating the python population. In recent years, political tensions have subsided, and 95% of the military forces have been withdrawn from the islands, allowing pythons and other wildlife populations to recover 40 years after being extirpated. Increased python predation on small livestock has recently become a nuisance to farmers, villagers, and aboriginal people, who are unaware of old python records from the island and therefore regard them as invasive pests. In order to clarify the origin of this python population, mitochondrial cytochrome b and COI genes were sequenced from 33 pythons from Kinmen, Fuzhou Zoo (China) and pet traders (Vietnam). Molecular phylogeny and haplotype networks showed a close relationship between Kinmen and Chinese populations, rejecting the speculation from local people that pythons were introduced from Southeast Asia. Population expansion of the python in recent years might be relevant to the retreat of the army, illustrating a novel case of demilitarized zone protecting wildlife. Since Kinmen Island has a similar latitude and climatic factors as the Everglades National Park in southern Florida (USA), where the Burmese python has become established, ecological and physiological research on the Kinmen population has potential value for comparative studies of habitat selection and niche modeling. Research on this new population could facilitate development of management plans for the Kinmen Island population and possibly help with managing the invasive population in the Florida Everglades.

Can settlement in natal-like habitat explain maladaptive habitat selection?

The study of habitat selection has long been influenced by the ideal free model, which maintains that young adults settle in habitat according to its inherent quality and the density of conspecifics within it. The model has gained support in recent years from the finding that conspecifics produce cues inadvertently that help prebreeders locate good habitat. Yet abundant evidence shows that animals often fail to occupy habitats that ecologists have identified as those of highest quality, leading to the conclusion that young animals settle on breeding spaces by means not widely understood. Here, we report that a phenomenon virtually unknown in nature, natal habitat preference induction (NHPI), is a strong predictor of territory settlement in both male and female common loons (Gavia immer). NHPI causes young animals to settle on natal-like breeding spaces, but not necessarily those that maximize reproductive success. If widespread, NHPI might explain apparently maladaptive habitat settlement.