Autocorrelation-informed home range estimation: a review and practical guide

Abstract

1. Modern tracking devices allow for the collection of high-volume animal tracking data at improved sampling rates over VHF radiotelemetry. Home range estimation is a key output from these tracking datasets, but the inherent properties of animal movement can lead traditional statistical methods to under- or overestimate home range areas. 2. The Autocorrelated Kernel Density Estimation (AKDE) family of estimators were designed to be statistically efficient while explicitly dealing with the complexities of modern movement data: autocorrelation, small sample sizes, and missing or irregularly sampled data. Although each of these estimators has been described in separate technical papers, here we review how these estimators work and provide a user-friendly guide on how they may be combined to reduce multiple biases simultaneously. 3. We describe the magnitude of the improvements offered by these estimators and their impact on home range area estimates, using both empirical case studies and simulations, contrasting their computational costs.4. Finally, we provide guidelines for researchers to choose among alternative estimators and an R script to facilitate the application and interpretation of AKDE home range estimates.

Lots of movement, little progress: a review of reptile home range literature

“…There is a growing body of work demonstrating the versatility of newer analytical methods (Noonan et al, 2019;Silva et al, 2020;Silva et al, 2021), and how they can be applied to the coarser resolution radio-telemetry data and the particulars of reptile movement (e.g., zero-inflated step lengths arising from long and frequent periods when the animal is stationary; Averill-Murray, Fleming & Riedle, 2020;Hromada et al, 2020;Silva et al, 2020). Reptile spatial ecology so far has largely failed to capitalise on the wealth of analytical options available, namely integrating movement information explicitly into estimations of space-use.…”

“…Unlike traditional estimation methods (KDEs and MCPs), movement-based area estimation models do not operate under the assumptions breached by tracking data (independence of points) and guard better against underand overestimation (Fleming & Calabrese, 2017;Silva et al, 2020). One of the common solutions to autocorrelation is the thinning of data; this procedure is inherently wasteful and inefficient, defeating the purpose of collecting high temporal-resolution data and reducing the biological relevance of telemetry datasets (Fleming et al, 2015;Calabrese et al, 2021). With low temporal-resolution data (typical of VHF data), analytic approaches will not necessarily reveal the correct home range patterns and need to be applied with caution; in these cases, it may be necessary to reconsider the research questions or re-evaluate study design for additional data collection.…”

A stable home: Autocorrelated Kernel Density Estimated home ranges of the critically endangered Elongated Tortoise

“…AKDEs have also been demonstrated to be more robust in the face of data gaps (Averill-Murray, Fleming & Riedle, 2020;Fleming et al, 2018). The manual collection of radio-tracking data, compared to regimented or automated GPS collection methods, makes AKDE an excellent analysis method to help address unforeseen lapses in data collection (due to staffing limitations, equipment failure, or inclement weather) particularly with the weighted AKDE function in the ctmm package (Calabrese, Fleming & Gurarie, 2016;Silva et al, 2021).…”

“…We use the ctmm package (Calabrese, Fleming & Gurarie, 2016; to fit all movement models using the perturbative hybrid REML method (pHREML), and estimate weighted AKDE home range areas (Silva et al, 2021). We used AICc to select the best fitting movement model for each individual, and selected the 95% contour to represent their home range area.…”

“…Here we investigate home range sizes for the tortoise (Indotestudo elongata) in Sakaerat Environmental Research Station, Nakhon Ratchasima, Thailand. We use the more appropriate and comparable Autocorrelated Kernel Density Estimators (AKDEs) as suggested by (Averill-Murray, Fleming & Riedle, 2020;Silva et al, 2021). We assess radio-tracked male and female elongated tortoise movements for range residency (to determine home range estimation is viable), and then compare male and female home ranges with an albeit limited sample.…”