Methods for Locating African Lion Kills Using Global Positioning System Movement Data

ABSTRACT Knowledge of the range, behavior, and feeding habits of large carnivores is fundamental to their successful conservation. Traditionally, the best method to obtain feeding data is through continuous observation, which is not always feasible. Reliable automated methods are needed to obtain sample sizes sufficient for statistical inference. Identification of large carnivore kill sites using Global Positioning System (GPS) data is gaining popularity. We assessed performance of generalized linear regression models (GLM) versus classification trees (CT) in a multipredator, multiprey African savanna ecosystem. We applied GLMs and CTs to various combinations of distance-traveled data, cluster durations, and environmental factors to predict occurrence of 234 female African lion (Panthera leo) kill sites from 1,477 investigated GPS clusters. Ratio of distance moved 24 hours before versus 24 hours after a cluster was the most important predictor variable in both GLM and CT analysis. In all cases, GLMs outperformed our cost-complexity-pruned CTs in their discriminative ability to separate kill from nonkill sites. Generalized linear models provided a good framework for kill-site identification that incorporates a hierarchal ordering of cluster investigation and measures to assess trade-offs between classification accuracy and time constraints. Implementation of GLMs within an adaptive sampling framework can considerably increase efficiency of locating kill sites, providing a cost-effective method for increasing sample sizes of kill data.     

Screening Global Positioning System Location Data for Errors Using Animal Movement Characteristics

Abstract: Animal locations estimated by Global Positioning System (GPS) inherently contain errors. Screening procedures used to remove large positional errors often trade data accuracy for data loss. We developed a simple screening method that identifies locations arising from unrealistic movement patterns. When applied to a large data set of moose (Alces alces) locations, our method identified virtually all known errors with minimal loss of data. Thus, our method for screening GPS data improves the quality of data sets and increases the value of such data for research and management.     

Differential prey use by male and female cougars in Washington

Male and female predators are often assumed to have the same effects on prey. Because of differences in body size and behavior, however, male and female predators may use different species, sexes, and ages of prey, which could have important implications for wildlife conservation and management. We tested for differential prey use by male and female cougars (Puma concolor) from 2003 to 2008 in Washington State. We predicted that male cougars would kill a greater proportion of larger and older prey (i.e., adult elk [Cervus elaphus]), whereas females would kill smaller and younger prey (i.e., elk calves, mule deer [Odocoileus hemionus]). We marked cougars with Global Positioning System (GPS) radio collars and investigated 436 predation sites. We located prey remains at 345 sites from 9 male and 9 female cougars. We detected 184 mule deer, 142 elk, and 17 remains from 4 other species. We used log-linear modeling to detect differences in species and age of prey killed among cougar reproductive classes. Solitary females and females with dependent offspring killed more mule deer than elk (143 vs. 83, P < 0.01), whereas males killed more elk than mule deer (59 vs. 41, P < 0.01). Proportionately, males killed 4 times more adult elk than did females (24% vs. 6% of kills) and females killed 2 times more adult mule deer than did males (26% vs. 15% of kills). Managers should consider the effects of sex of predator in conservation and management of ungulates, particularly when managing for sensitive species. © 2011 The Wildlife Society.     

Screening Radiolocation Datasets for Movement Strategies With Time Series Segmentation

ABSTRACT Classical home range analysis is tailored to meet requirements of data with few points per individual with relatively large intervals between observations. The swift rise in Global Positioning System (GPS)-based studies requires the development of new analytical approaches because GPS data allow for more detailed analysis in time and space. The amount of data derived from GPS studies enhances the potential to more accurately separate movement strategies. We present a general, simple, conceptual approach to using large movement datasets to automatically screen and delimit spatial and temporal home ranges of individuals and movement strategies using time series segmentation. We used GPS data for moose (Alces alces) from a boreal Swedish population as an example. We tested predictions that our screening method could separate seasonal migration from dispersal and nomadic strategies by the movement profile, which includes several dimensions. Our analysis showed that broad strategies were detected using our simple analytical approach, which speeds up use of GPS data for management and research because the method can be used to calculate more objective spatial and temporal activity ranges in relation to movement strategies. Our examples illustrate the importance of using the time stamp on location data in describing home ranges and movements.     

Assessment of Elk Highway Permeability by Using Global Positioning System Telemetry

Abstract: Highways have significant direct and indirect impact on natural ecosystems, including wildlife barrier and fragmentation effects, resulting in diminished habitat connectivity and highway permeability. We used Global Positioning System (GPS) telemetry to assess Rocky Mountain elk (Cervus elaphus nelsoni) permeability across a 30-km stretch of highway in central Arizona, USA, currently being reconstructed with 11 wildlife underpasses, 6 bridges, and associated ungulate-proof fencing. The highway was reconstructed in phases, allowing for comparison of highway crossing and passage rates during various stages of reconstruction. We instrumented 33 elk (25 F, 8 M) with GPS receiver collars May 2002 to April 2004. Our collars accrued 101,506 GPS fixes with 45% occurring within 1 km of the highway. Nearly 2 times the proportion of fixes occurred within 1 km of the highway compared with random. We think elk were attracted to the highway corridor by riparian—meadow foraging habitats that were 7 times more concentrated within the 1-km zone around the highway compared with the mean proportion within elk use areas encompassing all GPS fixes. Elk crossed the highway 3,057 times; crossing frequency and distribution along the highway were aggregated compared with random. Crossing frequency within 0.16-km highway segments was negatively associated with the distance to riparian—meadow habitats (r_s = -0.714, n = 190, P < 0.001). Mean observed crossing frequency (92.6 ± 23.5 [SE] crossings/elk) was lower than random (149.6 ± 27.6 crossings/elk). Females crossed 4.5 times as frequently as males. Highway permeability among reconstruction classes was assessed using passage rates (ratio of highway crossings to approaches); our overall mean passage rate was 0.67 ± 0.08 crossings per approach. The mean passage rate for elk crossing the highway section where reconstruction was completed (0.43 ± 0.15 crossings/approach) was half that of sections under reconstruction and control sections combined (0.86 ± 0.09 crossings/approach). Permeability was jointly influenced by the size of the widened highway and associated vehicular traffic on all lanes. Crossing frequency was used to delineate where ungulate-proof fencing yielded maximum benefit in intercepting and funneling crossing elk toward underpasses, promoting highway safety. Use of passage rates provides a quantitative measure to assess permeability, conduct future pre- and postconstruction comparisons, and to develop mitigation strategies to minimize highway impacts to wildlife.     

Effects of Global Positioning System Collar Weight on Zebra Behavior and Location Error

ABSTRACT Global Positioning System (GPS) collars are increasingly being used to study fine-scale patterns of animal behavior. Previous studies on GPS collars have tried to determine the causes of location error without attempting to investigate whether the accuracy of fixes provides a correspondingly accurate measure of the animal's natural behavior. When comparing 2 types of GPS collar, we found a significant effect of collar weight and fit on the rate of travel of plains zebra (Equus burchelli antiquorum) females in the Makgadikgadi, Botswana. Although both types of collar were well within accepted norms of collar weight, the slightly heavier collars (0.6% of total body mass [TBM]) reduced rate of travel by >50% when foraging compared with the collar that was 0.4% of TBM. Collar effect was activity specific, particularly interfering with grazing behavior; the effect was less noticeable when zebras crossed larger interpatch distances. We highlight that small differences in collar weight or fit can affect specific behaviors, limiting the extrapolation of fine-scaled GPS data. This has important implications for wildlife biologists, who hitherto have assumed that collars within accepted weight limits have little or no effect on animal movement parameters.     

Florida Panther Habitat Selection Analysis of Concurrent GPS and VHF Telemetry Data

Abstract: Florida panthers (Puma concolor coryi) are listed as an endangered subspecies in the United States and they exist in a single Florida population with <100 individuals; all known reproduction occurs south of Lake Okeechobee. Habitat loss is the biggest threat to this small population and previous studies of habitat selection have relied on very high frequency (VHF) telemetry data collected almost exclusively during diurnal periods. We investigated habitat selection of 12 panthers in the northern portion of the breeding range using 1) Global Positioning System (GPS) telemetry data collected during nocturnal and diurnal periods and 2) VHF telemetry data collected only during diurnal periods. Analysis of both types of telemetry data yielded similar results as panthers selected upland (P < 0.001) and wetland (P < 0.001) forested habitat types. Our results indicated that forests are the habitats selected by panthers and generally support the current United States Fish and Wildlife Service panther habitat ranking system. We suggest that future studies with greater numbers of panthers should investigate panther habitat selection using GPS telemetry data collected throughout the range of the Florida panther and with location attempts scheduled more evenly across the diel period. Global Positioning System radiocollars were effective at obtaining previously unavailable nocturnal telemetry data on panthers; however, we recommend that panther researchers continue to collect VHF telemetry data until acquisition rates and durability of GPS collars improve. (JOURNAL OF WILDLIFE MANAGEMENT 72(3):633–639; 2008)     

Relationship between daily body temperature and activity patterns of free-ranging feral cats in Australia

Daily rhythms of body temperature (T_b) and activity (distance travelled) of eight free-ranging feral cats (Felis catus) were recorded via implanted body temperature loggers in conjunction with Global Positioning System technology (GPS-radio collars), over a period of 14 days. The calculation of distance travelled (as a measure for activity) from GPS-data points proved to be efficient to quantify the relationship of both variables under field conditions. Analysis of covariance (ANCOVA) was employed to disentangle the relative effects of activity and time of day on T_b. Most variance in T_b was explained by time of day (with distinctly higher T_b at night), while considerably less variation was explained by distance travelled. Most importantly, the correlation between distance travelled and T_b was significantly stronger during daytime than at night. Indeed, night-time T_b showed little fluctuation at all. Taken together, the results suggest that the circadian T_b rhythm is primarily generated by an endogenous oscillator and that high T_b during night-time are not driven by high(er) nocturnal activity.     

Long-Distance Dispersal of a Rescued Wolf From the Northern Apennines to the Western Alps

ABSTRACT By using Global Positioning System technology, we documented the long-distance dispersal of a wolf (Canis lupus) from the northern Apennines in Italy to the western Alps in France. This is the first report of long-distance dispersal of wolves in the human-dominated landscapes of southern Europe, providing conclusive evidence that the expanding wolf population in the Alps originates from the Apennine source population through natural recolonization. By crossing 4 major 4-lane highways, agricultural areas, and several regional and provincial jurisdictions, the dispersal trajectory of wolf M15 revealed a single, narrow linkage connecting the Apennine and the Alpine wolf populations. This connectivity should be ensured to allow a moderate gene flow between the 2 populations and counteract potential bottleneck effects and reduced genetic variability of the Alpine wolf population. The case we report provides an example of how hard data can be effective in mitigating public controversies originating from the natural expansion and recolonization processes of large carnivore populations. In addition, by highlighting the connectivity between these 2 transboundary wolf populations, we suggest that documenting long-distance dispersal is particularly critical to support population-based, transboundary management programs.     

Implications of ignoring telemetry error on inference in wildlife resource use models

Global Positioning System (GPS) and very high frequency (VHF) telemetry data redefined the examination of wildlife resource use. Researchers collar animals, relocate those animals over time, and utilize the estimated locations to infer resource use and build predictive models. Precision of these estimated wildlife locations, however, influences the reliability of point-based models with accuracy depending on the interaction between mean telemetry error and how habitat characteristics are mapped (categorical raster resolution and patch size). Telemetry data often foster the assumption that locational error can be ignored without biasing study results. We evaluated the effects of mean telemetry error and categorical raster resolution on the correct characterization of patch use when locational error is ignored. We found that our ability to accurately attribute patch type to an estimated telemetry location improved nonlinearly as patch size increased and mean telemetry error decreased. Furthermore, the exact shape of these relationships was directly influenced by categorical raster resolution. Accuracy ranged from 100% (200-ha patch size, 1- to 5-m telemetry error) to 46% (0.5-ha patch size, 56- to 60-m telemetry error) for 10 m resolution rasters. Accuracy ranged from 99% (200-ha patch size, 1- to 5-m telemetry error) to 57% (0.5-ha patch size, 56- to 60-m telemetry error) for 30-m resolution rasters. When covariate rasters were less resolute (30 m vs. 10 m) estimates for the ignore technique were more accurate at smaller patch sizes. Hence, both fine resolution (10 m) covariate rasters and small patch sizes increased probability of patch misidentification. Our results help frame the scope of ecological inference made from point-based wildlife resource use models. For instance, to make ecological inferences with 90% accuracy at small patch sizes (≤5 ha) mean telemetry error ≤5 m is required for 10-m resolution categorical rasters. To achieve the same inference on 30-m resolution categorical rasters, mean telemetry error ≤10 m is required. We encourage wildlife professionals creating point-based models to assess whether reasonable estimates of resource use can be expected given their telemetry error, covariate raster resolution, and range of patch sizes. © 2011 The Wildlife Society.     

Performance of Spread Spectrum Global Positioning System Collars on Grizzly and Black Bears

ABSTRACT Global Positioning System (GPS) telemetry is a prevalent tool now used in the study of large mammals. Global Positioning Systems either store the data on board the collar or contain a remote-transfer system that allows for data recovery at more frequent intervals. Spread spectrum (S-S) technology is a new mode of data transfer designed to overcome interference problems associated with narrow-band very high frequency and ultra high frequency data-transfer systems. We evaluated performance of S-S GPS radiocollars deployed on grizzly (Ursus arctos) and black bears (U. americanus). We also evaluated variables that influenced GPS fix success rates, with particular focus on animal activity, time of year, and temperature. The S-S GPS collars performed to our expectations and met study objectives; we did not experience any major problems with the data-transfer system. We observed varying rates of fix success that were directly related to recorded activity counts. Using logistic regression, we verified that activity counts were a reasonable measure of resting or feeding-traveling in both bear species. Our results showed that 73% and 79% of missed fixes, respectively, occurred when we predicted black and grizzly bears to be resting. Temperatures measured in the canister of the collar were not correlated with air temperature, suggesting posture and activity influenced canister temperature. Both measures of temperature were predictive of fix success. We did not find that fix success was related to body morphology (i.e., neck circumference, mass, and chest girth), fix interval, position of the GPS antenna relative to the sky, or sex of the bear. We conclude that fix success for both species is strongly related to activity patterns and time of year. Activity counters appear to be a reasonable measure of this behavior, and we recommend researchers consider including an activity-count system when deploying GPS collars. We also recommend researchers explore building separate models of habitat selection based upon categories of activity to account for bias in fix success associated with bear behavior.     

Comparing Ground Telemetry and Global Positioning System Methods to Determine Cougar Kill Rates

ABSTRACT We assessed whether use of 2 methods, intensive very high frequency (VHF) radiotelemetry and Global Positioning System (GPS) cluster sampling, yielded similar estimates of cougar (Puma concolor) kill rates in Yellowstone National Park, 1998–2005. We additionally determined biases (underestimation or overestimation of rates) resulting from each method. We used modeling to evaluate what characteristics of clusters best predicted a kill versus no kill and further evaluated which predictor(s) minimized effort and the number of missed kills. We conducted 16 VHF ground predation sequences resulting in 37 kill intervals (KIs) and 21 GPS sequences resulting in 84 KIs on 6 solitary adult females, 4 maternal females, and 5 adult males. Kill rates (days/kill and biomass [kg] killed/day) did not differ between VHF and GPS predation sampling methods for maternal females, solitary adult females, and adult males. Sixteen of 142 (11.3%) kills detected via GPS clusters were missed through VHF ground-based sampling, and the kill rate was underestimated by an average of 5.2 (95% CI = 3.8–6.6) days/kill over all cougar social classes. Five of 142 (3.5%) kills identified by GPS cluster sampling were incorrectly identified as the focal individual's kill from scavenging, and the kill rate was overestimated within the adult male social class by an average of 5.8 (95% CI = 3.0–8.5) days/ungulate kill. The number of nights (locations between 2000 hours and 0500 hours) a cougar spent at a cluster was the most efficient variable at predicting predation, minimizing the missed kills, and minimizing number of extra clusters that needed to be searched. In Yellowstone National Park, where competing carnivores displaced cougars from their kills, it was necessary to search extra sites where a kill may not have been present to ensure we did not miss small, ungulate prey kills or kills with displacement. Using predictions from models to assign unvisited clusters as no kill, small prey kill, or large prey kill can bias downward the number of kills a cougar made and bias upward kills made by competitors that displace cougars or scavenge cougar kills. Our findings emphasize that field visitation is crucial in determining displacement and scavenging events that can result in biases when using GPS cluster methods in multicarnivore systems.     

Linking alternative food sources to winter habitat selection of herbivores in overbrowsed landscapes

During winter, ungulates in boreal forests must cope with high energetic costs related to locomotion in deep snow and reduced forage abundance and quality. At high density, ungulates face additional constraints, because heavy browsing reduces availability of woody browse, the main source of forage during winter. Under these severe conditions, large herbivores might forage on alternative food sources likely independent of browsing pressure, such as litterfall or windblown trees. We investigated the influence of alternative food sources on winter habitat selection, by studying female white-tailed deer (Odocoileus virginianus) living in 2 landscapes with contrasted browse abundance, recently logged and regenerated landscapes, in a population at high density and on a large island free of predators. We fitted 21 female white-tailed deer with Global Positioning System (GPS) collars and delineated winter home ranges and core areas. We measured snow conditions in different habitat categories and sampled vegetation in the core areas and in the rest of the home ranges to determine how forage abundance, protective cover, and snow conditions influenced habitat selection within the home range. In both landscapes, deer were less likely to use open habitat categories as snow accumulated on the ground. At a finer scale, deer inhabiting the regenerated landscape intensively used areas where balsam fir cover was intermediate with greater balsam fir browse density than in the rest of the home range. In the recently logged landscape, deer were more likely to be found near edges between clear-cuts and balsam fir stands and in areas where windblown balsam fir trees were present; the latter being the most influential variable. Although balsam fir browse was sparse and mainly out of reach in this landscape, deer increased the use of areas where it was present. Our results offer novel insights into the resource selection processes of northern ungulates, as we showed that access to winter forage, such as woody browse and alternative food sources, depends on climatic conditions and stochastic events, such as abundant compacted snow or windthrows. To compensate for these scarce and unpredictable food supplies, deer selected habitat categories, but mostly areas within those habitat categories, where the likelihood of finding browse, litterfall, and windblown trees was greatest. © 2011 The Wildlife Society.     

Contrasting Activity Patterns of Sympatric and Allopatric Black and Grizzly Bears

ABSTRACT The distribution of grizzly (Ursus arctos) and American black bears (U. americanus) overlaps in western North America. Few studies have detailed activity patterns where the species are sympatric and no studies contrasted patterns where populations are both sympatric and allopatric. We contrasted activity patterns for sympatric black and grizzly bears and for black bears allopatric to grizzly bears, how human influences altered patterns, and rates of grizzly-black bear predation. Activity patterns differed between black bear populations, with those sympatric to grizzly bears more day-active. Activity patterns of black bears allopatric with grizzly bears were similar to those of female grizzly bears; both were crepuscular and day-active. Male grizzly bears were crepuscular and night-active. Both species were more night-active and less day-active when ≤1 km from roads or developments. In our sympatric study area, 2 of 4 black bear mortalities were due to grizzly bear predation. Our results suggested patterns of activity that allowed for intra- and inter-species avoidance. National park management often results in convergence of locally high human densities in quality bear habitat. Our data provide additional understanding into how bears alter their activity patterns in response to other bears and humans and should help park managers minimize undesirable bear-human encounters when considering needs for temporal and spatial management of humans and human developments in bear habitats.     

Global Positioning System (GPS) location accuracy improvement due to selective availability removal

Global Positioning System (GPS) is an important new technology for spatio-temporal behaviour studies of animals. Differential correction improves location accuracy. Previously, it mostly removed partially the influence of Selective Availability (SA). SA was deactivated in May 2000. The aim of this study was to quantify the influence of SA cancellation on location accuracy of various GPS receivers. We tested the accuracy of locations obtained from non-differential and differential GPS animal collars before and after SA removal. We found a significant improvement in accuracy for both types of GPS collars. However, differential GPS still provides more accurate locations.