Restrictions on roaming Until the past century or so, the movement of wild animals was relatively unrestricted, and their travels contributed substantially to ecological processes. As humans have increasingly altered natural habitats, natural animal movements have been restricted. Tucker et al. examined GPS locations for more than 50 species. In general, animal movements were shorter in areas with high human impact, likely owing to changed behaviors and physical limitations. Besides affecting the species themselves, such changes could have wider effects by limiting the movement of nutrients and altering ecological interactions. Science , this issue p. 466

Home range estimation is routine practice in ecological research. While advances in animal tracking technology have increased our capacity to collect data to support home range analysis, these same advances have also resulted in increasingly autocorrelated data. Consequently, the question of which home range estimator to use on modern, highly autocorrelated tracking data remains open. This question is particularly relevant given that most estimators assume independently sampled data. Here, we provide a comprehensive evaluation of the effects of autocorrelation on home range estimation. We base our study on an extensive data set of GPS locations from 369 individuals representing 27 species distributed across five continents. We first assemble a broad array of home range estimators, including Kernel Density Estimation (KDE) with four bandwidth optimizers (Gaussian reference function, autocorrelated‐Gaussian reference function [AKDE], Silverman's rule of thumb, and least squares cross‐validation), Minimum Convex Polygon, and Local Convex Hull methods. Notably, all of these estimators except AKDE assume independent and identically distributed (IID) data. We then employ half‐sample cross‐validation to objectively quantify estimator performance, and the recently introduced effective sample size for home range area estimation (normalNfalse^area) to quantify the information content of each data set. We found that AKDE 95% area estimates were larger than conventional IID‐based estimates by a mean factor of 2. The median number of cross‐validated locations included in the hold‐out sets by AKDE 95% (or 50%) estimates was 95.3% (or 50.1%), confirming the larger AKDE ranges were appropriately selective at the specified quantile. Conversely, conventional estimates exhibited negative bias that increased with decreasing normalNfalse^area. To contextualize our empirical results, we performed a detailed simulation study to tease apart how sampling frequency, sampling duration, and the focal animal's movement conspire to affect range estimates. Paralleling our empirical results, the simulation study demonstrated that AKDE was generally more accurate than conventional methods, particularly for small normalNfalse^area. While 72% of the 369 empirical data sets had >1,000 total observations, only 4% had an normalNfalse^area >1,000, where 30% had an normalNfalse^area <30. In this frequently encountered scenario of small normalNfalse^area, AKDE was the only estimator capable of producing an accurate home range estimate on autocorrelated data.

Gray wolves (Canis lupus) in upper Michigan, USA, have been monitored since 1991 when breeding activity in mainland Michigan was documented for the first time since 1954. Based on winter track counts, the mean annual rate of increase in abundance was 19% from 1995 to 2002, with the population reaching an estimated 278 animals in 2002. Our objectives were to (1) increase the efficiency of wolf management in Michigan by evaluating alternative and less extensive sampling approaches for population estimation, and (2) evaluate habitat for wolves based on occupancy after a decade of recovery. For the first analysis, we created 22 discrete sampling units that cover upper Michigan, and we evaluated abundance estimates based on various sampling plans using known distribution and populations from the 2000-2002 winter track surveys. We evaluated each plan based on the precision, bias, and confidence interval coverage. A random sampling plan with regression estimator returned the most precise estimates, but a stratified sampling plan, using low, medium, and high wolf density strata had the greatest precision at lowest effort. For the habitat evaluation, we compared white-tailed (Odocoileus virginianus) deer density and road density between wolf pack locations from 1995 to 2001 to random locations outside of the current wolf range. We estimated white-tailed deer density by a spatial interpolation of pellet group counts. Our resource selection function indicated that probability of wolf occupation of an area was positively correlated with deer density, and it was relatively constant for road densities <0.4 km/km 2 but declined sharply at higher road densities. For areas habitable by wolves in upper Michigan, we predict a road density threshold of 0.7 km/km 2 and a deer density threshold of approximately 2.3-5.8 deer/km 2 . We believe that these results will aid managers who need to estimate wolf abundance and predict wolf distribution. JOURNAL OF WILDLIFE MANAGEMENT 69(4):1660-1669; 2005

Growth of ungulate populations is typically most sensitive to survival of neonates, which in turn is influenced by maternal nutritional condition and trade-offs in resource selection and avoidance of predators. We assessed whether resource use, multi-predator risk, maternal nutritional effects, hiding cover, or interactions among these variables best explained variation in daily survival of free-ranging neonatal white-tailed deer (Odocoileus virginianus) during their post-partum period (14 May–31 Aug) in Michigan, USA. We used Cox proportional hazards mixed-effects models to assess survival related to covariates of resource use, composite predation risk of 4 mammalian predators, fawn body mass at birth, winter weather, and vegetation growth phenology. Predation, particularly from coyotes (Canis latrans), was the leading cause of mortality; however, an additive model of non-ideal resource use and maternal nutritional effects explained 71% of the variation in survival. This relationship suggested that dams selected areas where fawns had poor resources, while greater predation in these areas led to additive mortalities beyond those related to resource use alone. Also, maternal nutritional effects suggested that severe winters resulted in dams producing smaller fawns, which decreased their likelihood of survival. Fawn resource use appeared to reflect dam avoidance of lowland forests with poor forage and greater use by wolves (C. lupus), their primary predator. While this strategy led to greater fawn mortality, particularly by coyotes, it likely promoted the life-long reproductive success of dams because many reached late-age (>10 years old) and could have produced multiple generations of fawns. Studies often link resource selection and survival of ungulates, but our results suggested that multiple factors can mediate that relationship, including multi-predator risk. We emphasize the importance of identifying interactions among biological and environmental factors when assessing survival of ungulates.