Use of GPS location clusters analysis to study predation, feeding, and maternal behavior of the Eurasian lynx

Abstract

Analysis of global positioning system (GPS) location clusters (GLCs) is becoming increasingly popular in studies of carnivore ecology. While promising, this application of GPS technology is still poorly developed for most species. We applied this method to study predation and maternal behavior of the Eurasian lynx (Lynx lynx) in Dinaric Mountains. Low population densities, rugged terrain, dense vegetation, and administrative borders make studies of this endangered population using traditional methods and a limited budget very challenging. We used geographic information system (GIS) and linear mixed effects models to understand the movement of lynx during the consumption process and denning period and estimate lynx kill rates. 99% of kills were found at GLCs longer than 30 h and with minimum two locations within 300 m. We confirmed 86% of potential kills and all potential dens that were searched for in the field. High success in predicting kill and den sites showed that Eurasian lynx is a suitable species for application of the GLC analysis methods. Comparison of field-confirmed kills with model predictions showed the possibility for remote estimation of approximate kill 2 rates in Eurasian lynx. Movements of lynx were primarily affected by daytime period, time since the last kill/den translocation, lynx category and their interactions. Based on the empirical data we programmed simulations of lynx movements and elaborated recommendations for more efficient field procedures and study designs (GPS schedules) for future studies. We believe that our findings and approach will also benefit studies of other species with similar behavior.

Spatial ecology of a herd of white-lipped peccaries (Tayassu pecari) in Belize using GPS telemetry: challenges and preliminary results

“…The numerous reports of successfully deployed GPS collars in different environmental settings convincingly illustrate the potential that this equipment holds for obtaining large quantities of high quality data in wildlife research (Edenius 1996;Eriksen et al 2011;Martins et al 2011;Harju et al 2013;Krofel et al 2013). However, the use of GPS collars often comes with unforeseen technical or data quality issues (Sager-Fradkin et al 2007;Matthews et al 2013).…”

Identification of kill sites from GPS clusters for jaguars (Panthera onca) in the southern Pantanal, Brazil

“…Similarly, Svoboda et al (2013) found that clusters with more locations increased their odds of finding a bobcat (Lynx rufus) kill site. Krofel et al (2013) used GPS cluster locations to find kill sites and den sites of Eurasian lynx (Lynx lynx), with 99% of the kill sites found at clusters longer than 30 h and with a minimum of two locations within 300 m. Our research is the first attempt to use spatial, temporal, and habitat characteristics of potential clusters to identify jaguar kill sites in South America. We found that the duration of time a jaguar spent at the cluster, the number of nights a jaguar spent at a cluster and the dispersion of the points at the cluster were the best predictors of whether a cluster was associated with a kill.…”

“…In addition, GPS collars can provide data on movement distances, which are useful for estimating population density when combined with remote camera trapping (Soisalo and Cavalcanti 2006), as well as information on denning behaviour (e.g. Krofel et al 2013). The only disadvantages of GPS collars may be the initial costs and the bias for large prey when the relocation acquisition schedule of the GPS collars is temporally far enough apart between locations that small prey items are completely consumed in that time interval, and therefore no cluster is identified for investigation in the field.…”

“…During early studies using GPS collars, Anderson and Lindzey (2003) demonstrated that GPS locations collected within a particular time-frame and distance from each other could be used to evaluate if clusters of locations were most likely associated with a kill. Since then, multiple studies on large carnivores have used GPS collar locations to determine which clusters are potential kill sites (e. g. Sand et al 2005;Webb et al 2008;Knopff et al 2009;Ruth et al 2010;Tambling et al 2010;Martins et al 2011;Pitman et al 2012;Krofel et al 2013;Miller et al 2013). Interestingly, as technology has advanced and the use of GPS collars has increased, the volume of locational data has also increased, creating a situation that is both beneficial and deleterious.…”

“…Interestingly, as technology has advanced and the use of GPS collars has increased, the volume of locational data has also increased, creating a situation that is both beneficial and deleterious. The large amount of data allows researchers to access locations at both spatially and temporally fine scales while at the same time generating too many clusters that can be realistically investigated in the field to determine if they are kill sites or sites associated with other behaviours not related to predation (e. g. Webb et al 2008;Knopff et al 2009;Ruth et al 2010;Krofel et al 2013).…”