Are kernels the mustard? Data from global positioning system (GPS) collars suggests problems for kernel home- range analyses with least-squares cross-validation

By Graham Hemson, P. Johnson, Andy South, R. Kenward, Ruth M. Ripley, David W. Macdonald. 2005

365 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/666e152639eaade5acc389413e215361528ee8b4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology

In Small Animals

Citations in the corpus

Spatial structure of Amur (Siberian) tigers (Panthera tigris altaica) on Sikhote-Alin Biosphere Zapovednik, Russia (2010, background): “Fixed kernel estimators have been criticized for a number of reasons (Getz et al. 2007; Hemson et al. 2005).”
Living in a landscape of fear: the impact of predation, resource availability and habitat structure on primate range use (2014, neutral): “This method was chosen over kernel density estimation owing to its superior convergence properties and ability to cope better with hard boundaries such as cliffs and rivers and clumped data points (Getz et al., 2007; Hemson et al., 2005; Ryan, Knechtel, & Getz, 2006; Silverman, 1986; Steury, McCarth”
Spatial ecology and conservation of seabirds facing global climate change: a review (2009, positive): “Furthermore , the pertinence of kernel analyses, which have also been routinely used to map the distribution of seabirds at sea, has been criticized (Hemson et al. 2005), and should likewise be reassessed and potentially replaced by more refined mapping techniques.”
Group territoriality and the benefits of sociality in the African lion, Panthera leo (2009, background): “Because of the potential sensitivity of kernel density estimates to sample size (Anderson 1982; Worton 1987, 1989; Harris et al. 1990; Hemson et al. 2005), we chose a 75% contour as a less biased estimate of territory area and location.”
Rigorous home range estimation with movement data: a new autocorrelated kernel density estimator. (2015, background): “Because any single data point will be strongly correlated with adjacent data points, leaving only one data point out of an estimate will not yield any substantial change, which will then be falsely attributed to fit quality (also, see Hemson et al. 2005, for other issues).”
Does the risk of encountering lions influence African herbivore behaviour at waterholes? (2009, methodology): “To build such landscape, we calculated the consecutive 10% isopleths kernel, using the reference smoothing factor href as recommended by Hemson et al. (2005), for each individual lion using Ranges 7 (version 0.811, South and Kenward 2006).”
Effects of Point Pattern Shape on Home-Range Estimates (2008, background): “Hemson et al. (2005) also argued that KDE is not as accurate of a home-range estimator as the literature suggests.”
Estimating Utilization Distributions with Kernel Versus Local Convex Hull Methods (2011, methodology): “For instance, UDs allow likelihood-based comparisons of alternative behavioral models (Horne and Garton 2006, Moorcroft et al. 2006).”
How key habitat features influence large terrestrial carnivore movements: waterholes and African lions in a semi-arid savanna of north-western Zimbabwe (2010, methodology): “Home ranges were calculated using the 95% probability contour of all locations distribution using the kernel density estimator, a method with recognized strengths (Worton 1989; Powell 2000), and the reference smoothing factor href (recommended by Hemson et al. 2005; Börger et al. 2006).”
What Is the Proper Methodto Delineate Home Range of anAnimal Using Today’s AdvancedGPS Telemetry Systems: The Initial Step (2011, neutral): “…analysis with uncorrelated locations, kernel-density estimation (KDE), has been lauded by some for use with GPS technology (Kie et al., 2010) while criticized by others for errors in proper bandwidth selection (Hemson et al., 2005) and violation of independence assumptions (Swihart & Slade 1985b).”
A Critical Review of Home Range Studies (2008, methodology): “LSCV search function employed had little influence on KDE estimates (but see Horne and Garton 2006). Jones et al. (1996) recommended the solve-the-equation and plug-in method for bandwidth selection. Rodgers and Carr (1998) recommended standardizing or scaling the location data before applying KDE, ”