Justin A. Gude

Prey Risk Allocation In A Grazing Ecosystem

Understanding the behaviorally mediated indirect effects of predators in ecosystems requires knowledge of predator-prey behavioral interactions. In predator-ungulate-plant systems, empirical research quantifying how predators affect ungulate group sizes and distribution, in the context of other influential variables, is particularly needed. The risk allocation hypothesis proposes that prey behavioral responses to predation risk depend on background frequencies of exposure to risk, and it can be used to make predictions about predator-ungulate-plant interactions. We determined non-predation variables that affect elk (Cervus elaphus) group sizes and distribution on a winter range in the Greater Yellowstone Ecosystem (GYE) using logistic and log-linear regression on surveys of 513 1-km2 areas conducted over two years. Employing model selection techniques, we evaluated risk allocation and other a priori hypotheses of elk group size and distributional responses to wolf (Canis lupus) predation risk while accounting for influential non-wolf-predation variables. We found little evidence that wolves affect elk group sizes, which were strongly influenced by habitat type and hunting by humans. Following predictions from the risk allocation hypothesis, wolves likely created a more dynamic elk distribution in areas that they frequently hunted, as elk tended to move following wolf encounters in those areas. This response should dilute elk foraging pressure on plant communities in areas where they are frequently hunted by wolves. We predict that this should decrease the spatial heterogeneity of elk impacts on grasslands in areas that wolves frequently hunt. We also predict that this should decrease browsing pressure on heavily browsed woody plant stands in certain areas, which is supported by recent research in the GYE.

Generalizing wolf effects across the Greater Yellowstone Area: a cautionary note

Abstract Although numerous authors are investigating indirect effects of wolf recovery, the most fundamental ecological impact of the Greater Yellowstone Area wolf reintroduction, the effects of wolf predation on ungulate populations, remains unclear. We report on a 5-year comparative study of wolf (Canis lupus)–elk (Cervus elaphus) dynamics on an elk herd in the headwaters of the Madison River within Yellowstone National Park and the lower Madison elk herd that winters 40 km downriver outside the Park. A resident pack became established on the Madison headwaters area in 1997 and grew to multiple packs totaling 30–40 animals by 2002. During winter 1999 emigrates from Yellowstone established a pack on the lower Madison area. However, poor recruitment and low adult survival limited wolf population growth, with the area supporting a single pack, never exceeding 5 animals. Wolf kill rates on the lower Madison area were approximately double that documented for the Madison headwaters area. Moderate kill rates in the Madison headwaters, combined with high wolf densities and modest elk densities, resulted in an estimated 20% of the elk population being killed during winter and projections for a declining elk population. In contrast, high kill rates on the lower Madison area, combined with low wolf densities and high elk densities, resulted in winter predation estimates not exceeding 4% of the elk population. We suspect this level of mortality will be of little biological significance with respect to elk population trajectory. These results suggest that the effects of wolf predation on elk populations differ substantially over relatively small spatial scales, depending on a complex suite of interacting factors. Thus, we caution against generalizing the effects of wolf restoration on elk dynamics from any single study and encourage collaborations to develop comparative predator–prey studies that improve our understanding of wolf–ungulate interactions and enhance conservation.

Determining Occurrence Dynamics when False Positives Occur: Estimating the Range Dynamics of Wolves from Public Survey Data

Large-scale presence-absence monitoring programs have great promise for many conservation applications. Their value can be limited by potential incorrect inferences owing to observational errors, especially when data are collected by the public. To combat this, previous analytical methods have focused on addressing non-detection from public survey data. Misclassification errors have received less attention but are also likely to be a common component of public surveys, as well as many other data types. We derive estimators for dynamic occupancy parameters (extinction and colonization), focusing on the case where certainty can be assumed for a subset of detections. We demonstrate how to simultaneously account for non-detection (false negatives) and misclassification (false positives) when estimating occurrence parameters for gray wolves in northern Montana from 2007–2010. Our primary data source for the analysis was observations by deer and elk hunters, reported as part of the state’s annual hunter survey. This data was supplemented with data from known locations of radio-collared wolves. We found that occupancy was relatively stable during the years of the study and wolves were largely restricted to the highest quality habitats in the study area. Transitions in the occupancy status of sites were rare, as occupied sites almost always remained occupied and unoccupied sites remained unoccupied. Failing to account for false positives led to over estimation of both the area inhabited by wolves and the frequency of turnover. The ability to properly account for both false negatives and false positives is an important step to improve inferences for conservation from large-scale public surveys. The approach we propose will improve our understanding of the status of wolf populations and is relevant to many other data types where false positives are a component of observations.

Relative influence of human harvest, carnivores, and weather on adult female elk survival across western North America

Summary 1. Well-informed management of harvested species requires understanding how changing ecological conditions affect demography and population dynamics, information that is lacking for many species. We have limited understanding of the relative influence of carnivores, harvest, weather and forage availability on elk Cervus elaphus demography, despite the ecological and economic importance of this species. We assessed adult female survival, a key vital rate for population dynamics, from 2746 radio-collared elk in 45 populations across western North America that experience wide variation in carnivore assemblage, harvest, weather and habitat conditions. 2. Proportional hazard analysis revealed that ‘baseline’ (i.e. not related to human factors) mortality was higher with very high winter precipitation, particularly in populations sympatric with wolves Canis lupus. Mortality may increase via nutritional stress and heightened vulnerability to predation in snowy winters. Baseline mortality was unrelated to puma Puma concolor presence, forest cover or summer forage productivity. 3. Cause-specific mortality analyses showed that wolves and all carnivore species combined had additive effects on baseline elk mortality, but only reduced survival by <2%. When human factors were included, ‘total’ adult mortality was solely related to harvest; the influence of native carnivores was compensatory. Annual total mortality rates were lowest in populations sympatric with both pumas and wolves because managers reduced female harvest in areas with abundant or diverse carnivores.

Changes in Elk Resource Selection and Distributions Associated With a Late-Season Elk Hunt

Abstract Changes in resource selection associated with human predation risk may alter elk distributions and availability for harvest. We used Global Positioning System data collected from telemetered female elk (Cervus elaphus) to evaluate effects of refuges (areas where hunting was prohibited), spatial variation in hunting risk, and landscape attributes on resource selection within an established Greater Yellowstone Area, USA, winter range. We also evaluated elk distributions during and outside of a late-season hunting period. Refuge areas and landscape attributes such as habitat type and snow water equivalents (SWE) affected resource selection. Elk selection for flat grasslands increased as SWE increased, likely because these areas were windswept, leaving grasses exposed for foraging. Elk distributions differed during hunting and no-hunting periods. During the hunting period, elk shifted to privately owned refuge areas and the estimated odds of elk occupying refuge areas more than doubled. Risk-driven changes in resource selection resulted in reduced availability of elk for harvest. Elk selection for areas where hunting is prohibited presents a challenge for resource managers that use hunting as a tool for managing populations and influences grazing patterns on private ranchlands.

Anthropogenic mortality, intraspecific competition, and prey availability influence territory sizes of wolves in Montana

Abstract Territoriality in animals is of both theoretical and conservation interest. Animals are territorial when benefits of exclusive access to a limiting resource outweigh costs of maintaining and defending it. The size of territories can be considered a function of ecological factors that affect this benefit–cost ratio. Previous research has shown that territory sizes for wolves (Canis lupus) are largely determined by available biomass of prey, and possibly pack size and density of neighboring wolf packs, but has not been interpreted in a benefit–cost framework. Such a framework is relevant for wolves living in the Northern Rocky Mountains where conflicts with humans increase mortality, thereby potentially increasing costs of being territorial and using prey resources located near humans. We estimated territory sizes for 38 wolf packs in Montana from 2008 to 2009 using 90% adaptive kernels. We then created generalized linear models (GLMs) representing combinations of ecological factors hypothesized to affect the territory sizes of wolf packs. Our top GLM, which had good model fit (R2  =  0.68, P < 0.0005), suggested that territory sizes of wolves in Montana were positively related to terrain ruggedness, lethal controls, and human density and negatively related to number of surrounding packs relative to the size of the territory. We found that the top GLM successfully predicted territory sizes (R2  =  0.53, P < 0.0005) using a jackknife approach. Our study shows that territory sizes of group-living carnivores are influenced by not only intraspecific competition and availability of limiting resources, but also by anthropogenic threats to the groups survival, which could have important consequences where these territorial carnivores come into conflict with humans.

Estimation of Successful Breeding Pairs for Wolves in the Northern Rocky Mountains, USA

Abstract Under the Endangered Species Act, documenting recovery and federally mandated population levels of wolves (Canis lupus) in the Northern Rocky Mountains (NRM) requires monitoring wolf packs that successfully recruit young. United States Fish and Wildlife Service regulations define successful breeding pairs as packs estimated to contain an adult male and female, accompanied by ≥2 pups on 31 December of a given year. Monitoring successful breeding pairs will become more difficult following proposed delisting of NRM wolves; alternatives to historically intensive methods, appropriate to the different ecological and regulatory context following delisting, are required. Because pack size is easier to monitor than pack composition, we estimated probability a pack would contain a successful breeding pair based on its size for wolf populations inhabiting 6 areas in the NRM. We also evaluated the extent to which differences in demography of wolves and levels of human-caused mortality among the areas influenced the probability of packs of different sizes would contain successful breeding pairs. Probability curves differed among analysis areas, depending primarily on levels of human-caused mortality, secondarily on annual population growth rate, and little on annual population density. Probabilities that packs contained successful breeding pairs were more uniformly distributed across pack sizes in areas with low levels of human mortality and stable populations. Large packs in areas with high levels of human-caused mortality and high annual growth rates had relatively high probabilities of containing breeding pairs whereas those for small packs were relatively low. Our approach can be used by managers to estimate number of successful breeding pairs in a population where number of packs and their sizes are known. Following delisting of NRM wolves, human-caused mortality is likely to increase, resulting in more small packs with low probabilities of containing breeding pairs. Differing contributions of packs to wolf population growth based on their size suggests monitoring successful breeding pairs will provide more accurate insights into population dynamics of wolves than will monitoring number of packs or individuals only.

Internal Validation of Predictive Logistic Regression Models for Decision-Making in Wildlife Management

Predictive logistic regression models are commonly used to make informed decisions related to wildlife management and conservation, such as predicting favourable wildlife habitat for land conservation objectives and predicting vital rates for use in population models. Frequently, models are developed for use in the same population from which sample data were obtained, and thus, they are intended for internal use within the same population. Before predictions from logistic regression models are used to make management decisions, predictive ability should be validated. We describe a process for conducting an internal model validation, and we illustrate the process of internal validation using logistic regression models for predicting the number of successfully breeding wolf packs in six areas in the US northern Rocky Mountains. We start by defining the major components of accuracy for binary predictions as calibration and discrimination, and we describe methods for quantifying the calibration and discrimination abilities of a logistic regression model. We also describe methods for correcting problems of calibration and future predictive accuracy in a logistic regression model. We then show how bootstrap simulations can be used to obtain unbiased estimates of prediction accuracy when models are calibrated and evaluated within the same population from which they were developed. We also show how bootstrapping can be used to assess coverage rates and recalibrate the endpoints of confidence intervals for predictions from a logistic regression model, to achieve nominal coverage rates. Using the data on successfully breeding wolf packs in the northern Rocky Mountains, we validate that predictions from a model developed with data specific to each of six analysis areas are better calibrated to each population than a global model developed using all data simultaneously. We then use shrinkage of model coefficients to improve calibration and future predictive accuracy for the area-specific model, and recalibrate confidence interval endpoints to provide better coverage properties. Following this validation, managers can be confident that logistic regression predictions will be reliable in this situation, and thus that management decisions will be based on accurate predictions.

Temporal validation of an estimator for successful breeding pairs of wolves Canis lupus in the U.S. northern Rocky Mountains

Abstract Model-based predictors derived from historical data are rarely evaluated before they are used to draw inferences. We performed a temporal validation, (i.e. assessed the performance of a predictive model using data collected from the same population after the model was developed) of a statistical predictor for the number of successful breeding pairs of wolves Canis lupus in the northern Rocky Mountains (NRM). We predicted the number of successful breeding pairs, β, in Idaho, Montana and Wyoming based on the distribution of pack sizes observed through monitoring in 2006 and 2007 (β̂), and compared these estimates to the minimum number of successful breeding pairs, βMIN, observed through intensive monitoring. βMIN was consistently included within the 95% confidence intervals of β̂ for all states in both years (except for Idaho in 2007), generally following the pattern β̂L (lower 95% prediction interval for β̂) < β̂MIN < β̂. This evaluation of β̂ estimates for 2006 and 2007 suggest it will be a robust model-based method for predicting successful breeding pairs of NRM wolves in the future, provided influences other than those modeled in β̂ (e.g. disease outbreak, severe winter) do not have a strong effect on wolf populations. Managers can use β̂ models with added confidence as part of their post-delisting monitoring of wolves in NRM.