Showing papers in "Journal of Wildlife Management in 2006"
TL;DR: A new validation method is concluded for evaluating predictive performance of a RSF and for assessing if the model deviates from being proportional to the probability of use of a resource unit.
Abstract: Applications of logistic regression in a used–unused design in wildlife habitat studies often suffer from asymmetry of errors: used resource units (landscape locations) are known with certainty, whereas unused resource units might be observed to be used with greater sampling intensity. More appropriate might be to use logistic regression to estimate a resource selection function (RSF) tied to a use–availability design based on independent samples drawn from used and available resource units. We review the theoretical motivation for RSFs and show that sample “contamination” and the exponential form commonly assumed for the RSF are not concerns, contrary to recent statements by Keating and Cherry (2004; Use and interpretation of logistic regression in habitat-selection studies. Journal of Wildlife Management 68:774–789). To do this, we re-derive the use–availability likelihood and show that it can be maximized by logistic regression software. We then consider 2 case studies that illustrate our find...
696 citations
TL;DR: It is illustrated that resource selection models are part of a broader collection of statistical models called weighted distributions and recommend some promising areas for future development.
Abstract: We review 87 articles published in the Journal of Wildlife Management from 2000 to 2004 to assess the current state of practice in the design and analysis of resource selection studies. Articles were classified into 4 study designs. In design 1, data are collected at the population level because individual animals are not identified. Individual animal selection may be assessed in designs 2 and 3. In design 2, use by each animal is recorded, but availability (or nonuse) is measured only at the population level. Use and availability (or unused) are measured for each animal in design 3. In design 4, resource use is measured multiple times for each animal, and availability (or nonuse) is measured for each use location. Thus, use and availability measures are paired for each use in design 4. The 4 study designs were used about equally in the articles reviewed. The most commonly used statistical analyses were logistic regression (40%) and compositional analysis (25%). We illustrate 4 problem areas in resource selection analyses: pooling of relocation data across animals with differing numbers of relocations, analyzing paired data as though they were independent, tests that do not control experiment wise error rates, and modeling observations as if they were independent when temporal or spatial correlations occurs in the data. Statistical models that allow for variation in individual animal selection rather than pooling are recommended to improve error estimation in population-level selection. Some researchers did not select appropriate statistical analyses for paired data, or their analyses were not well described. Researchers using one-resource-at-a-time procedures often did not control the experiment wise error rate, so simultaneous inference procedures and multivariate assessments of selection are suggested. The time interval between animal relocations was often relatively short, but existing analyses for temporally or spatially correlated data were not used. For studies that used logistic regression, we identified the data type employed: single sample, case control (used-unused), use-availability, or paired use-availability. It was not always clear whether studies intended to compare use to nonuse or use to availability. Despite the popularity of compositional analysis, we do not recommend it for multiple relocation data when use of one or more resources is low. We illustrate that resource selection models are part of a broader collection of statistical models called weighted distributions and recommend some promising areas for future development.
649 citations
TL;DR: In this paper, the authors compare the performance of the scaling least square cross-validation (LSCV) and reference approaches to plug-in and solve-the-equation (STE) bandwidth methods.
Abstract: In studies of animal space use, researchers often use kernel-based techniques for estimating the size of an animal's home range and its utilization distribution from radiotracking data. However, the kernel estimator is highly sensitive to the bandwidth value used. Previous ecological studies recommended least-squares cross-validation (LSCV) as the default bandwidth selection method, but some statisticians consider this technique inferior to newer methods. We used simulations to compare the performance of the scaling LSCV and reference approaches to plug-in and solve-the-equation (STE) bandwidth methods. We generated samples of 20, 50, and 150 points from mixtures of 2, 4, and 16 bivariate normal distributions. We selected the ranges of potential variances for these distributions to create 4 distribution types with varied levels of clumping to simulate the diversity of location patterns expected from radiotracking data. For most distribution types, plug-in and STE methods performed as well or better than LSCV in % absolute error of home-range size estimates and overlap of estimated and true utilization distributions. Although the relative differences usually were small, the plug-in and STE approaches provide good alternatives to LSCV. However, LSCV performed better with distribution types composed entirely of tight clumps of points. The reference bandwidth performed poorly for most distributions. Surprisingly, it often had the lowest absolute error at outer contours for distributions consisting of a single very tight cluster surrounded by more dispersed points. Although our results demonstrate the utility of plug-in and STE approaches, no method was best across all distributions. Rather, choice of a bandwidth selection method may vary depending on the study goals, sample size, and patterns of space use by the study species. In general, we recommend plug-in and STE approaches for estimating relatively smooth outer contours. The LSCV approach is better at identifying tight clumps, including areas of peak use, although risk of LSCV failure also increases when a distribution has a very tight cluster of points. When planning to use kernel methods, researchers should consider these factors to make preliminary decisions about the bandwidth method expected to be most appropriate in their study.
325 citations
TL;DR: It is illustrated how misclassifying units as unused may lead to incorrect conclusions about resource use, and how recently developed occupancy models can be utilized within the resource-selection context to improve conclusions by explicitly accounting for detection probability.
Abstract: Resource-selection probability functions and occupancy models are powerful methods of identifying areas within a landscape that are highly used by a species. One common design/analysis method for estimation of a resource-selection probability function is to classify a sample of units as used or unused and estimate the probability of use as a function of independent variables using, for example, logistic regression. This method requires that resource units are correctly classified as unused (i.e., the species is never undetected in a used unit), or that the probability of misclassification is the same for all units. In this paper, I explore these issues, illustrating how misclassifying units as unused may lead to incorrect conclusions about resource use. I also show how recently developed occupancy models can be utilized within the resource-selection context to improve conclusions by explicitly accounting for detection probability. These models require that multiple surveys be conducted at each of a sample of resource units within a relatively short timeframe, but given the growing evidence from simulation studies and field data, I recommend that such procedures should be incorporated into studies of resource use.
322 citations
TL;DR: In this paper, the authors examined winter habitat selection patterns of adult female mule deer before and during the first 3 years of development in a natural gas field in western Wyoming and observed shifts in their distribution as development progressed toward less-preferred and presumably less-suitable habitats.
Abstract: Increased levels of natural gas exploration, development, and production across the Intermountain West have created a variety of concerns for mule deer (Odocoileus hemionus) populations, including direct habitat loss to road and well-pad construction and indirect habitat losses that may occur if deer use declines near roads or well pads. We examined winter habitat selection patterns of adult female mule deer before and during the first 3 years of development in a natural gas field in western Wyoming. We used global positioning system (GPS) locations collected from a sample of adult female mule deer to model relative frequency or probability of use as a function of habitat variables. Model coefficients and predictive maps suggested mule deer were less likely to occupy areas in close proximity to well pads than those farther away. Changes in habitat selection appeared to be immediate (i.e., year 1 of development), and no evidence of well-pad acclimation occurred through the course of the study; rather, mule deer selected areas farther from well pads as development progressed. Lower predicted probabilities of use within 2.7 to 3.7 km of well pads suggested indirect habitat losses may be substantially larger than direct habitat losses. Additionally, some areas classified as high probability of use by mule deer before gas field development changed to areas of low use following development, and others originally classified as low probability of use were used more frequently as the field developed. If areas with high probability of use before development were those preferred by the deer, observed shifts in their distribution as development progressed were toward less-preferred and presumably less-suitable habitats.
320 citations
TL;DR: Choosing an appropriate smoothing parameter is critical when using kernel methods to estimate animal home ranges, and this study provides useful guidelines when making this decision.
Abstract: Fixed kernel density analysis with least squares cross-validation (LSCVh) choice of the smoothing parameter is currently recommended for home-range estimation. However, LSCVh has several drawbacks, including high variability, a tendency to undersmooth data, and multiple local minima in the LSCVh function. An alternative to LSCVh is likelihood cross-validation (CVh). We used computer simulations to compare estimated home ranges using fixed kernel density with CVh and LSCVh to true underlying distributions. Likelihood cross-validation generally performed better than LSCVh, producing estimates with better fit and less variability, and it was especially beneficial at sample sizes <˜50. Because CVh is based on minimizing the Kullback-Leibler distance and LSCVh the integrated squared error, for each of these measures of discrepancy, we discussed their foundation and general use, statistical properties as they relate to home-range analysis, and the biological or practical interpretation of these statist...
273 citations
TL;DR: The rationale for estimation of RSFs based on UDs, advice about computing UDs and RSFs, and their use in resource selection studies are described, as well as 3 case studies that demonstrate use of UDs inresource selection modeling.
Abstract: Often resource selection functions (RSFs) are developed by comparing resource attributes of used sites to unused or available ones. We present alternative approaches to the analysis of resource selection based on the utilization distribution (UD). Our objectives are to describe the rationale for estimation of RSFs based on UDs, offer advice about computing UDs and RSFs, and illustrate their use in resource selection studies. We discuss the 3 main factors that should be considered when using kernel UD-based estimates of space use: selection of bandwidth values, sample size versus precision of estimates, and UD shape and complexity. We present 3 case studies that demonstrate use of UDs in resource selection modeling. The first example demonstrates the general case of RSF estimation that uses multiple regression adjusted for spatial autocorrelation to relate UD estimates (i.e., the probability density function) to resource attributes. A second example, involving Poisson regression with an offset term, is presented as an alternative for modeling the relative frequency, or probability of use, within defined habitat units. This procedure uses the relative frequency of locations within a habitat unit as a surrogate of the UD and requires relatively fewer user-defined options in the modeling of resource selection. Last, we illustrate how the UD can also be used to enhance univariate resource selection analyses, such as compositional analysis, in cases where animals use their range nonrandomly. The UD helps overcome several common shortcomings of some other analytical techniques by treating the animal as the primary sampling unit, summarizing use in a continuous and probabilistic manner, and relying on the pattern of animal space use rather than using individual sampling points. However, several drawbacks are apparent when using the UD in resource selection analyses. Choice of UD estimator is important and sensitive to sample size and user-defined options, such as bandwidth and software selection. Extensions to these procedures could consider behavioral-based approaches and alternative techniques to estimate the UD directly.
272 citations
TL;DR: It is believed that site occupancy is a useful state variable and suggest that monitoring programs for mammals using occupancy data consider detectability prior to making inferences about species distributions or population change.
Abstract: Large-scale, multispecies monitoring programs are widely used to assess changes in wildlife populations but they often assume constant detectability when documenting species occurrence. This assumption is rarely met in practice because animal populations vary across time and space. As a result, detectability of a species can be influenced by a number of physical, biological, or anthropogenic factors (e.g., weather, seasonality, topography, biological rhythms, sampling methods). To evaluate some of these influences, we estimated site occupancy rates using species-specific detection probabilities for meso- and large terrestrial mammal species on Cape Cod, Massachusetts, USA. We used model selection to assess the influence of different sampling methods and major environmental factors on our ability to detect individual species. Remote cameras detected the most species (9), followed by cubby boxes (7) and hair traps (4) over a 13-month period. Estimated site occupancy rates were similar among sampling methods for most species when detection probabilities exceeded 0.15, but we question estimates obtained from methods with detection probabilities between 0.05 and 0.15, and we consider methods with lower probabilities unacceptable for occupancy estimation and inference. Estimated detection probabilities can be used to accommodate variation in sampling methods, which allows for comparison of monitoring programs using different protocols. Vegetation and seasonality produced species-specific differences in detectability and occupancy, but differences were not consistent within or among species, which suggests that our results should be considered in the context of local habitat features and life history traits for the target species. We believe that site occupancy is a useful state variable and suggest that monitoring programs for mammals using occupancy data consider detectability prior to making inferences about species distributions or population change.
202 citations
TL;DR: In this paper, the authors review the 7 major classes of mark-recapture models that investigators can use to estimate apparent survival and other parameters from live-encounter data.
Abstract: Estimation of demographic parameters is central to research questions in wildlife management, conservation, and evolutionary ecology. I review the 7 major classes of mark–recapture models that investigators can use to estimate apparent survival and other parameters from live-encounter data. Return rates are the product of 4 probabilities: true survival (S), site fidelity (F), site propensity (δ), and true detection (p*). Cormack-Jolly-Seber (CJS) models improve upon return rates by separating apparent survival (ϕ = S × F) from the probability of encounter (p = δ × p*). The main drawback to mark–recapture models based on live-encounter data is that the complement of apparent survival (1 − ϕ) includes losses to mortality and to permanent emigration, and these 2 ecological processes are difficult to disentangle. Advanced mark–recapture models require additional sampling effort but estimate apparent survival with greater precision and less bias, and they also offer estimates of other useful demograph...
200 citations
TL;DR: In this paper, the authors investigated the influence of patch size and landscape composition on density and nesting success of three grassland passerines, after controlling for local-scale vegetation structure, climate, and when analyzing nest success.
Abstract: Current management recommendations for grassland birds in North America emphasize providing large patches of grassland habitat within landscapes that have few forest or shrubland areas. These Bird Conservation Areas are being proposed under the assumption that large patches of habitat in treeless landscapes will maintain viable populations of grassland birds. This assumption requires that patch size and landscape features affect density and nesting success of grassland birds, and that these effects are consistent among years and regions and across focal species. However, these assumptions have not yet been validated for grassland birds, and the relative importance of local vegetation structure, patch size, and landscape composition on grassland bird populations is not well known. In addition, factors influencing grassland bird nesting success have been investigated mostly in small-scale and short-duration studies. To develop management guidelines for grassland birds, we tested the spatial and temporal repeatability of the influence of patch size and landscape composition on density and nesting success of 3 grassland passerines, after controlling for local-scale vegetation structure, climate, and—when analyzing nest success—bird density. We conducted our study during 4 years (1998–2001) in 44 study plots that were set up in 3 regions of the northern tallgrass prairie in Minnesota and North Dakota, USA. In these study plots we measured density and nesting success of clay-colored sparrows (Spizella pallida), Savannah sparrows (Passerculus sandwichensis), and bobolinks (Dolichonyx oryzivorus). Statistical models indicated that density was influenced by patch size, landscape, region, and local vegetation structure more so than by local vegetation structure alone. Both magnitude and direction of the response of density to patch size varied among regions, years, and species. In contrast, the direction of landscape effects was consistent among regions, years, and between Savannah sparrows and bobolinks. In each species, this landscape effect was independent of patch size. Nesting success was not clearly influenced by patch size or landscape composition, and none of the factors that influenced avian density also influenced nesting success in any of the 3 species. General statements on “optimal habitat” for grassland birds should therefore be viewed cautiously. Instead, long-term studies in different regions as well as a deeper understanding of the local system are needed to determine which factors are most important for grassland birds in a particular area.
177 citations
TL;DR: In this article, the authors used the generalized Horvitz-Thompson estimator, based on the overall detection probability for each individual dugong, to generate population estimates and developed a new simulation-based method for estimating standard errors and confidence intervals.
Abstract: The probability of detecting an animal in a sampled area during a survey consists of 2 components: 1) the probability of an animal being available for detection (availability), which can be highly variable in heterogeneous environments; and 2) the probability of an animal being detected, conditional on its being available for detection (perception). Many surveys only estimate the latter probability because modeling the availability process requires information collected external to the survey. We illustrate estimation of both probabilities in an application to aerial surveys of dugongs (Dugong dugon) in Northern Australian coastal waters where water clarity varies greatly over relatively small spatial scales. Using artificial dugong models and timed depth recorders deployed on 15 wild dugongs to obtain dive profiles, we carried out experiments to determine zones of detectability for dugongs at the range of depths, turbidities, and sea states that spanned the environmental heterogeneity encountered on dugong surveys. Resulting probability estimates were heterogeneous and dependent on the measured conditions. To estimate perception probability, we used a tandem team of 2 observers on either side of the aircraft. This permitted fitting generalized Lincoln–Petersen models with Program MARK. We then used the generalized Horvitz–Thompson estimator, based on the overall detection probability for each individual dugong, to generate population estimates. We also developed a new simulation-based method for estimating standard errors and confidence intervals. We contrast absolute abundance estimates of dugongs in the Torres Strait and Northern Great Barrier Reef regions using both the new and original approaches (Marsh and Sinclair 1989a). For Torres Strait, the new method produced a substantially smaller estimate (11,956 vs. 14,106 dugongs) and a very much smaller standard error (1,189 vs. 2,314 dugongs), whereas the new method produced slightly larger estimates (mean 9,855 vs. 9,193 dugongs, standard error 1,184 vs. 917 dugongs) for the Northern Great Barrier Reef survey.
TL;DR: In this article, the authors considered cause-specific mortality from a modern biostatistical perspective and proposed an appropriate hazard-based representation as a competing risks problem, which leads to the general solution of causespecific mortality as the cumulative incidence function (CIF) and describes the appropriate generalization of the fully nonparametric staggered-entry Kaplan-meier survival e...
Abstract: Estimating cause-specific mortality is often of central importance for understanding the dynamics of wildlife populations. Despite such importance, methodology for estimating and analyzing cause-specific mortality has received little attention in wildlife ecology during the past 20 years. The issue of analyzing cause-specific, mutually exclusive events in time is not unique to wildlife. In fact, this general problem has received substantial attention in human biomedical applications within the context of biostatistical survival analysis. Here, we consider cause-specific mortality from a modern biostatistical perspective. This requires carefully defining what we mean by cause-specific mortality and then providing an appropriate hazard-based representation as a competing risks problem. This leads to the general solution of cause-specific mortality as the cumulative incidence function (CIF). We describe the appropriate generalization of the fully nonparametric staggered-entry Kaplan–Meier survival e...
TL;DR: Advantages of using MARK to estimate the parameters of the various multistate models include flexibility of model specification to include group, time, and individual covariates, estimation of variance components, model averaging of parameter estimates, and Bayesian parameter estimation using Markov chain Monte Carlo procedures on the logit scale.
Abstract: Program MARK provides >100 models for the estimation of population parameters from mark–encounter data. The multistate model of Brownie et al. (1993) and Hestbeck et al. (1991) allows animals to move between states with a probability of transition. The simplest multistate model is an extension of the Cormack–Jolly–Seber (CJS) live recapture model. Parameters estimated are state-specific survival rates and encounter probabilities and transition probabilities between states. The multistate model provides a valuable framework to evaluate important ecological questions. For example, estimation of state-specific survival and transition probabilities between the biological states of breeders and nonbreeders allows estimation of the cost of reproduction. Transitions between physical states, such as spatial areas, provide estimates needed for meta-population models. The basic multistate model uses only live recaptures, but 3 extensions are included in MARK. A multistate model with live and dead encounter...
TL;DR: This article used more than 10 years of data on buffalo herds in a Geographic Information System (GIS) of Klaserie Private Nature Reserve (KPNR) to examine ranging behavior and habitat selection at multiple temporal and geographic scales.
Abstract: We used more than 10 years of data on buffalo herds in a Geographic Information System (GIS) of Klaserie Private Nature Reserve (KPNR) to examine ranging behavior and habitat selection at multiple temporal and geographic scales. We compared 3 methods of empirical home range estimation: minimum convex polygons (MCP); a fixed-kernel method; and a new local nearest-neighbor convex-hull construction method (LoCoH). For 3 herds over 5 years (1995–2000), the southern herd (SH) had the largest range, the focal study herd (FH) had the intermediate range, and the northern herd (NH) had the smallest range. The LoCoH method best-described the ranges because it accommodated user knowledge of known physical barriers, such as fences, whereas the MCP and kernel methods overestimated ranges. Short-term ranges of the FH over 9 years reveal that buffalo travel farther and range wider in the dry season than the wet. Habitat selection analyses on broad vegetation categories showed preference for Acacia shrub veld an...
TL;DR: In this article, the authors identify important local, stand, and landscape factors influencing bat habitat use in northwestern South Carolina, USA, and use an information theoretic approach to determine the variables that best predicted use by bats.
Abstract: Knowledge and understanding of bat habitat associations and the responses of bats to forest management are critical for effective bat conservation and management. Few studies have been conducted on bat habitat use in the southeast, despite the high number of endangered and sensitive species in the region. Our objective was to identify important local, stand, and landscape factors influencing bat habitat use in northwestern South Carolina, USA. We hypothesized that use would be greatest 1) at points with relatively sparse vegetation, 2) in early successional and mature stands, and 3) at points close to streams. We also predicted that species would exhibit different patterns of habitat use based on morphology. We placed Anabat II bat detectors at points located in 3 forest types and 3 age classes to record bats from May–August 2004 and 2005. We used an information theoretic approach to determine the variables that best predicted use by bats. Vegetation density at the sample point was the best predictor of overall bat presence in 2004. In 2005 vegetation density and distance to the closest road were the best predictors of overall bat use; the model containing age class also had good support. Bats were more likely to be recorded at points with sparse vegetation, farther from roads, and in early successional stands. Vegetation density was also the best predictor of habitat use by big brown bats (Eptesicus fuscus) and red bats (Lasiurus borealis); both species were far more likely to be recorded at points with sparse vegetation at the sample point. Eastern pipistrelles (Pipistrellus subflavus) were also more likely to be recorded at points with sparse vegetation and in early successional stands. The best predictors of northern long-eared bat (Myotis septentrionalis) habitat use were vegetation density and age class. Northern long-eared bats were more likely to be recorded at points with sparse vegetation and in mature stands. Our results suggest that early successional habitats and small openings and gaps within forest stands provide suitable commuting and foraging bat habitat in northwestern South Carolina. However, mature forests are also important for some species. Forest management practices that provide a variety of age classes across the landscape and that create gaps and openings within mid- and late-successional stands will likely provide suitable habitat for bats in the mountains of South Carolina.
TL;DR: Telemetry-based survival estimation will benefit from increased emphasis on modeling approaches designed to elucidate survival determinants in complex systems, combined with more rigorous attention to basic assumptions and study design limitations.
Abstract: Survival estimation is an important aspect of population ecology and conservation biology, and radiotelemetry is a major tool for assessing factors influencing survival time in free-ranging birds and mammals. Despite the advantage of telemetry in providing extensive and continuous survival information, telemetry-based survival estimates can be biased or imprecise when methods are misused. Simple cumulative survival estimators like the Heisey and Fuller and Kaplan-Meier methods have underlying assumptions and sampling requirements that commonly remain unverified by researchers. Telemetry studies often limit survival analysis to simple univariate tests that do not consider the range of factors potentially influencing mortality risk in free-ranging animals. Continuous-time modeling approaches like Cox Proportional Hazards or Anderson–Gill methods, or their discrete-time analogues, are superior because they are robust to a range of study design limitations and can handle multiple categorical or conti...
TL;DR: The National Wildlife Strike Database for Civil Aviation in the United States contained 38,961 reports of aircraft collisions with birds (bird strikes) from 1990-2004 in which the report indicated the height above ground level (AGL) as discussed by the authors.
Abstract: The National Wildlife Strike Database for Civil Aviation in the United States contained 38,961 reports of aircraft collisions with birds (bird strikes) from 1990–2004 in which the report indicated the height above ground level (AGL). I analyzed these strike reports to determine the distribution of all strikes and those strikes causing substantial damage to aircraft by height. For the 26% of strikes above 500 feet (152 m) AGL (n = 10,143), a simple negative exponential model, with height as the independent variable, explained 99% of the variation in number of bird strikes per 1,000-foot (305-m) interval. Strikes declined consistently by 32% every 1,000 feet from 501–20,500 feet (153–6,248 m). For strikes at ≤500 feet, passerines, gulls and terns, pigeons and doves, and raptors were the identified species groups most frequently struck. For strikes at >500 feet, waterfowl, gulls and terns, passerines, and vultures were the species groups most frequently struck. For strikes that resulted in substanti...
TL;DR: Impact of forest fragmentation on bat communities is described and information that can be used when integrating forest-management practices into bat conservation is provided.
Abstract: Changes in structure and arrangement of forests may influence the distribution of bat communities by affecting roosting and foraging habitat. Using Anabat bat detectors, we determined presence of bat species at 316 sample plots in southeastern Missouri, USA, through qualitative identification of echolocation calls collected. We used maximum-likelihood estimation techniques incorporating detection probabilities into estimation of site occupancy by species of bats. We compared a priori models at 2 geographic scales using information theoretic methods. At the local-site scale, eastern pipistrelle (Pipistrellus subflavus) and red bat (Lasiurus borealis) occupancy was most influenced by structural characteristics of forested areas, whereas Indiana bats (Myotis sodalis) were influenced most by density of large-diameter snags that could provide roosting habitat. At the landscape scale, occupancy of Indiana bats was directly related to amount of nonforested land cover. Northern long-eared bat (M. septentrionalis) occupancy was inversely related to edge. These data describe implications of forest fragmentation and provide information that can be used when integrating forest-management practices into bat conservation.
TL;DR: Study of species- and gender-specific aspects of summer range extent and stand selection in northern long-eared bats and little brown bats in the Greater Fundy Ecosystem, New Brunswick, Canada suggested that understanding gender effects is crucial for accurate characterization of forest bat habitats.
Abstract: To understand bat biology and appreciate their dependence on and role within forested ecosystems, the biological resolution at which studies are directed must elucidate species and gender patterns. We studied species- and gender-specific aspects of summer range extent and stand selection in northern long-eared bats (Myotis septentrionalis) and little brown bats (M. lucifugus) in the Greater Fundy Ecosystem, New Brunswick, Canada, using trapping, radiotelemetry, and ultrasonic monitoring. Our results suggested that this 2-species system is comprised of 4 ecologically distinct groups with respect to site selection and range extent for roosting and foraging. All bats exhibited an affinity to specific roosting areas. Myotis septentrionalis roosted and foraged in the forest interior. The roosting and foraging areas for females were 6.1 times and 3.4 times larger, respectively, than for males. Both genders foraged in site types in proportion to their availability. Myotis lucifugus females roosted in buildings outside the core study area, and those captured in the forested landscape were transients. Compared to male and female M. septentrionalis, male M. lucifugus had intermediate-sized roosting areas but the largest foraging areas. Water sites were selected during foraging more than expected. Bat foraging activity, measured acoustically at 8 site types, was greatest at lakes and least above the forest canopy. Male M. lucifugus activity levels were positively associated with temperature and the amount of mature coniferous forest and water within 1 km of the sampling site, and they were negatively associated with the amount of mature deciduous forest within 1 km of the site. Our results suggested that understanding gender effects is crucial for accurate characterization of forest bat habitats. Studies of bats that combine data for genders, species, or guilds may produce spurious results and may be of minimal value for, or actually hinder, bat conservation and management programs.
TL;DR: The combined effects of hunters killing prime-aged females (2–9 yr old), wolves killing calves, and predation by other predators has the potential to limit the elk population in the future.
Abstract: We compared selection of northern Yellowstone elk (Cervus elaphus) by hunters in the Gardiner Late Hunt and northern Yellowstone wolves (Canis lupus) with regard to sex, age, and impacts to recruitment. We compared harvest data from 1996–2001 with wolf-killed elk data from 1995–2001. We assessed the effects of hunting and wolf predation on reproductive female elk by constructing a life table and calculating reproductive values for females in the northern Yellowstone herd. We devised an index of total reproductive impact to measure impacts to calf production due to hunting and wolf predation. The age classes of female elk selected by wolves and hunters were significantly different. Hunters selected a large proportion of female elk with the greatest reproductive values, whereas wolves selected a large proportion of elk calves and older females with low reproductive values. The mean age of adult females killed by hunters throughout the study period was 6.5 years, whereas the mean age of adult females killed by wolves was 13.9 years. Hunting exerted a greater total reproductive impact on the herd than wolf predation. The combined effects of hunters killing prime-aged females (2–9 yr old), wolves killing calves, and predation by other predators has the potential to limit the elk population in the future. Yellowstone is unique in this regard because multiple predators that occur sympatrically, including hunters, wolves, grizzly bears (Ursus arctos), black bears (Ursus americanus), cougars (Felis concolor), and coyotes (Canis latrans), all prey on elk. Using an Adaptive Harvest Management process the known female elk harvest during the Gardiner Late Hunt has been reduced by 72% from 2,221 elk in 1997 to 620 elk in 2004. In the future, hunting harvest levels may be reduced further to partially offset elk losses to wolves, other predators, and environmental factors.
TL;DR: In this paper, the authors compared the spatial ecology of Bobcats and gray foxes between urban and rural zones of a national park in northern California, USA, and found that adult female bobcat home ranges were smaller in the urban zone, and core areas were both smaller and overlapped more.
Abstract: Urbanization threatens the persistence of many wildlife populations, particularly those of wide-ranging and low-density species such as mammalian carnivores. Effective conservation of carnivore populations requires an understanding of the impacts of adjacent urbanization on carnivores in reserves. I compared the spatial ecology of bobcats (Lynx rufus) and gray foxes (Urocyon cinereoargenteus) between urban and rural zones of a national park in northern California, USA. In the urban zone, gray foxes used the entire landscape from interior natural areas across the park edges and into the neighboring developed areas, although fox core areas were always within the park. Bobcats never entered development, and radiocollared adult female bobcats maintained home ranges in the interior of the park, far from the urban edge. Bobcats appeared to avoid crossing paved roads, while foxes crossed roads regularly. For adult female bobcats, home ranges were smaller in the urban zone, and core areas were both smaller and overlapped more. Home range size and overlap did not differ between zones for gray foxes. Bobcats seem to be more affected by the proximity of urbanization than foxes, perhaps because of differences in diet and social structure. The more flexible use of the landscape by foxes may give them access to increased resources and habitat, but also may expose them to more human-associated risks. If female bobcats are more sensitive to urbanization, this sensitivity could affect the long-term viability of bobcat populations in urban areas. Knowledge of how bobcats and gray foxes use the landscape in urban areas will allow more effective conservation and improved coexistence with these widespread carnivores by helping to predict where and why conservation or management issues may occur.
TL;DR: This paper used multistage sampling (MSS) to estimate waste-rice abundance during September-December 2000-2002, and found that the abundance of waste rice in late autumn differed between harvester types (i.e., conventional > stripper header).
Abstract: Flooded rice fields are important foraging habitats for waterfowl in the lower Mississippi Alluvial Valley (MAV). Waste rice previously was abundant in late autumn (140–492 kg/ha), but early planting and harvest dates in recent years may have increased losses of waste rice during autumn before waterfowl arrive. Research in Mississippi rice fields revealed waste-rice abundance decreased 79–99% during autumns 1995–1996 (Manley et al. 2004). To determine if this trend existed throughout the MAV, we used multistage sampling (MSS) to estimate waste-rice abundance during September–December 2000–2002. Averaged over years, mean abundance of waste rice decreased 71% between harvest (x = 271.0 kg/ha, CV = 13% n = 3 years) and late autumn (x = 78.4 kg/ha, CV = 15% n = 3). Among 15 models formulated to explain variation in rice abundance among fields and across years, the best model indicated abundance of waste rice in late autumn differed between harvester types (i.e., conventional > stripper header) and ...
TL;DR: This paper argues that perhaps the most informative currency of use would be increased risk to fitness accepted by an animal, and discusses numerous factors and criteria that should be considered when delineating the area an animal might use during a period of interest.
Abstract: Resource availability and resource use are 2 key concepts in studies of resource selection. Although equal accessibility to resources is one component defining resource availability, we rarely know what restricts access to resources. Consequently for spatially distributed resources, the animal's use of space in association with the occurrence of resources is a frequent basis for inferring resource use and testing for resource selection. For many resources, occurrence can be defined for the population or the individual animal and requires that researchers specify the spatial extent of resources an animal might use during the time interval of study (e.g., the “choice set”). Often the occurrence of resources is defined at multiple scales, which facilitates understanding hierarchical selection patterns. We discuss numerous factors and criteria that should be considered when delineating the area an animal might use during a period of interest. New analytical approaches to resource selection, including resource utilization functions (RUF) and discrete choice modeling, help address some of the issues of defining availability and dealing with the behaviors associated with resource use. A currency of use is a measure of the investment made by an animal in securing resources, avoiding loss of resources, or otherwise optimizing fitness. Common currencies used by researchers include time spent or distance traveled in a cover type, use of event sites (e.g., nest sites, roosting sites, den sites), or amounts of different kinds of foods consumed. Less common, but potentially highly informative, are such currencies as energy expended or predation risk or other risk incurred. Simulation of animal movements interspersed with diel resting periods, through habitat types with activity-dependent energy expenditure and habitat-specific predation risk showed that choice of a currency of use strongly influences inferences about habitat selection. We argue that perhaps the most informative currency of use would be increased risk to fitness accepted by an animal. Although fairly simple conceptually, such application of risk assessment faces formidable empirical challenges and is a worthy goal for the next generation of researchers of animal resource selection.
TL;DR: This work generalizes the classic discrete-choice model to the situation in which multiple choices are made from 1 or more choice sets, and only 1 random sample from each choice set is available, and shows that logistic regression provides a good approximation to discrete choice models if the expected number of uses is much <1 for all units.
Abstract: Discrete-choice models are a powerful and flexible method for studying habitat selection, in part because they allow resource availability to change at every choice. Here, we consider application of discrete-choice models to data typically collected in wildlife science because different discrete-choice data are usually collected in other disciplines. We generalize the classic discrete-choice model to the situation in which multiple choices are made from 1 or more choice sets, and only 1 random sample from each choice set is available. We discuss analysis using 1) logistic regression, 2) maximum likelihood when choices are made with replacement, 3) maximum likelihood when the temporal order of selection is known, and 4) maximum likelihood when the order of selection is unknown. We show that 1) provides a good approximation to discrete choice models if the expected number of uses is much 41 years old with high levels of hardwoods adjacent to stands 6–20 yrs old.
TL;DR: In this paper, the authors used active acoustical sampling (Anabat II) to survey foraging habitat relationships of bats on the Savannah River Site (SRS) in the upper Coastal Plain of South Carolina.
Abstract: During 2001, we used active acoustical sampling (Anabat II) to survey foraging habitat relationships of bats on the Savannah River Site (SRS) in the upper Coastal Plain of South Carolina. Using an a priori information-theoretic approach, we conducted logistic regression analysis to examine presence of individual bat species relative to a suite of microhabitat, stand, and landscape-level features such as forest structural metrics, forest type, proximity to riparian zones and Carolina bay wetlands, insect abundance, and weather. There was considerable empirical support to suggest that the majority of the activity of bats across most of the 6 species occurred at smaller, stand-level habitat scales that combine measures of habitat clutter (e.g., declining forest canopy cover and basal area), proximity to riparian zones, and insect abundance. Accordingly, we hypothesized that most foraging habitat relationships were more local than landscape across this relatively large area for generalist species of bats. The southeastern myotis (Myotis austroriparius) was the partial exception, as its presence was linked to proximity of Carolina bays (best-approximating model) and bottomland hardwood communities (other models with empirical support). Efforts at SRS to promote open longleaf pine (Pinus palustris) and loblolly pine (P. taeda) savanna conditions and to actively restore degraded Carolina bay wetlands will be beneficial to bats. Accordingly, our results should provide managers better insight for crafting guidelines for bat habitat conservation that could be linked to widely accepted land management and environmental restoration practices for the region.
TL;DR: The findings suggest that cougars in the Pacific Northwest are currently declining, and are recommend reduced levels of exploitation, particularly for adult females, continuous monitoring, and collaborative efforts of managers from adjacent states and provinces.
Abstract: Increasing reports of human/cougar conflicts may suggest that cougars are increasing in the Pacific Northwest. We determined minimum relative densities and average fecundity, survival, and growth rate of an apparently increasing cougar population in northeastern Washington, USA; northern Idaho, USA; and southern British Columbia, Canada, from 1998 to 2003. Minimum relative densities declined from 1.47 cougars/100 km2 to 0.85 cougars/100 km2. We estimated average litter size at 2.53 kittens, interbirth interval at 18 months, proportion of reproductively successful females at 75%, and age at first parturition at 18 months for a maternity rate of 1.27 kittens/adult female/yr. Average survival rate for all radiocollared cougars was 59%: 77% for adult females, 33% for adult males, 34% for yearlings, and 57% for kittens. Hunting accounted for 92% of mortalities of radiocollared cougars. The annual stochastic growth rate of this population was λ = 0.80 (95% CI = 0.11). Contrary to accepted belief, our f...
TL;DR: In this paper, the authors examined wolf-habitat relationships in the northern Rocky Mountains of the U.S. by relating landscape and habitat features found within wolf pack home ranges to those found in adjacent non-occupied areas.
Abstract: Gray wolf (Canis lupus) populations have persisted and expanded in northwest Montana since 1986, while reintroduction efforts in Idaho and Yellowstone have further bolstered the regional population. However, rigorous analysis of either the availability of wolf habitat in the entire region, or the specific habitat requirements of local wolves, has yet to be conducted. We examined wolf-habitat relationships in the northern Rocky Mountains of the U.S. by relating landscape/habitat features found within wolf pack home ranges (n = 56) to those found in adjacent non-occupied areas (n = 56). Logistic regression revealed that increased forest cover, lower human population density, higher elk density, and lower sheep density were the primary factors related to wolf occupation. Similar factors promoted wolf pack persistence. Further, our analysis indicated that relatively large tracts of suitable habitat remain unoccupied in the Rocky Mountains, suggesting that wolf populations likely will continue to increase in the region. Analysis of the habitat linkage between the 3 main wolf recovery areas indicates that populations in central Idaho and northwest Montana have higher connectivity than either of the 2 recovery areas to the Greater Yellowstone recovery area. Thus, for the northern Rocky Mountains to function as a metapopulation for wolves, it will be necessary that dispersal corridors to the Yellowstone ecosystem be established and conserved.
TL;DR: In this paper, a Bayesian random-effects model is presented to assess resource selection in an arctic coastal plain calving ground using radiolocation data from 76 adult female caribou (Rangifer tarandus) and calf pairs.
Abstract: Modeling the probability of use of land units characterized by discrete and continuous measures, we present a Bayesian random-effects model to assess resource selection. This model provides simultaneous estimation of both individual- and population-level selection. Deviance information criterion (DIC), a Bayesian alternative to AIC that is sample-size specific, is used for model selection. Aerial radiolocation data from 76 adult female caribou (Rangifer tarandus) and calf pairs during 1 year on an Arctic coastal plain calving ground were used to illustrate models and assess population-level selection of landscape attributes, as well as individual heterogeneity of selection. Landscape attributes included elevation, NDVI (a measure of forage greenness), and land cover-type classification. Results from the first of a 2-stage model-selection procedure indicated that there is substantial heterogeneity among cow-calf pairs with respect to selection of the landscape attributes. In the second stage, selection of models with heterogeneity included indicated that at the population-level, NDVI and land cover class were significant attributes for selection of different landscapes by pairs on the calving ground. Population-level selection coefficients indicate that the pairs generally select landscapes with higher levels of NDVI, but the relationship is quadratic. The highest rate of selection occurs at values of NDVI less than the maximum observed. Results for land cover-class selections coefficients indicate that wet sedge, moist sedge, herbaceous tussock tundra, and shrub tussock tundra are selected at approximately the same rate, while alpine and sparsely vegetated landscapes are selected at a lower rate. Furthermore, the variability in selection by individual caribou for moist sedge and sparsely vegetated landscapes is large relative to the variability in selection of other land cover types. The example analysis illustrates that, while sometimes computationally intense, a Bayesian hierarchical discrete-choice model for resource selection can provide managers with 2 components of population-level inference: average population selection and variability of selection. Both components are necessary to make sound management decisions based on animal selection.
TL;DR: In this article, the authors studied Red fox, raccoons, and striped skunks' use of levees and the edges associated with them within a wetland environment and found that the densities of these predators have increased throughout the Intermountain West as a consequence of human alterations in habitat.
Abstract: Red foxes (Vulpes vulpes), raccoons (Procyon lotor), and striped skunks (Mephitis mephitis) are found throughout the United States, wherever there is suitable denning habitat and food resources. Densities of these predators have increased throughout the Intermountain West as a consequence of human alterations in habitat. Within the Bear River Migratory Bird Refuge (hereafter, refuge), in northern Utah, USA, upland nesting habitat for ducks is limited to the levee banks and roadsides. Red foxes, raccoons, and striped skunks, which prey on upland nesting birds, are also abundant on the refuge. We studied red foxes, raccoons, and striped skunks' use of levees and the edges associated with them within a wetland environment. Red fox, raccoon, and striped skunk locations were negatively correlated with distance to the nearest dike (−0.78, −0.69, and −0.45, respectively). Animals incorporated more roads and/or levees into their home ranges than expected by chance (x = 2.6; Z < 0.001); incorporation of levees was greater during the dispersal season than the rearing season (P = 0.03). Skunk home ranges (average size, 3.0 km2) were oriented along roads and levees (P = 0.03), whereas raccoon (average size, 3.6 km2) and fox home ranges (average size, 3.5 km2) were not (P = 0.93, P = 0.13, respectively). Fox home ranges in the refuge were more oblong in shape than reported elsewhere (P = 0.03). However, home-range shapes of raccoons and striped skunks were similar to previous studies (P = 0.84, P = 0.97, respectively). The use of roads and levees within the refuge increases the possible travel distance and penetration of predators into wetland environments. This contributes to increased depredation of waterfowl nests and to decreased recruitment. Managers of similar areas might decrease depredation of waterfowl by disrupting the linear pattern of corridors, thereby decreasing the congestion of animal roads and levees. This would, then, decrease the encounter rates of predators and prey.
TL;DR: It is concluded that CWD is not randomly distributed among deer and that differential transmission among sex and age classes is likely driving the observed patterns in disease prevalence, and several possible nonlinear relationships between prevalence and age are discussed.
Abstract: Chronic wasting disease (CWD) is a fatal disease of white-tailed deer (Odocoileus virginianus) caused by transmissible protease-resistant prions. Since the discovery of CWD in southern Wisconsin in 2001, more than 20,000 deer have been removed from a >2,500-km2 disease eradication zone surrounding the three initial cases. Nearly all deer removed were tested for CWD infection and sex, age, and harvest location were recorded. Our analysis used data from a 310-km2 core study area where disease prevalence was higher than surrounding areas. We found no difference in harvest rates between CWD infected and noninfected deer. Our results show that the probability of infection increased with age and that adult males were more likely to be infected than adult females. Six fawns tested positive for CWD, five fawns from the core study area, including the youngest (5 months) free-ranging cervid to test positive. The increase in male prevalence with age is nearly twice the increase found in females. We concluded that CWD is not randomly distributed among deer and that differential transmission among sex and age classes is likely driving the observed patterns in disease prevalence. We discuss alternative hypotheses for CWD transmission and spread and, in addition, discuss several possible nonlinear relationships between prevalence and age. Understanding CWD transmission in free-ranging cervid populations will be essential to the development of strategies to manage this disease in areas where CWD is found, as well as for surveillance strategies in areas where CWD threatens to spread.