To conserve the mobility of species across changing land and seascapes, we must first understand how much space is necessary to maintain stable, interconnected populations. Home range estimation allows managers and policymakers to easily visualize the habitats most commonly used by species of conservation concern.


Location, occurrence, and range graphs

Figure 1: GPS location data (top panel) can be used to determine both where an animal might have traveled during observation (occurrence distribution) and predict where it might go in the future (range distribution).

Home range estimation presents several quantitative challenges and is prone to statistical biases that can lead to underestimation (Fig. 1). This can have negative impacts on conservation outcomes as it may result in conservation managers underestimating how much land should be protected or overestimating the number of animals that a region can sustain.​

Smithsonian scientists have overcome these biases by first accounting for the autocorrelation present in telemetry data. Autocorrelation is a statistic that contains the signature of the animal’s unknown movement path. As a result, autocorrelation-informed methods (autocorrelated kernel density estimation; AKDE) provide more reliable predictions of where animals may travel in the future as compared to traditional methods that depend on an assumption of unrelated locations (with no movement path connecting them; Fig. 2).


Home-range estimation maps

Figure 2: Conventional methods of estimating home-range size can result in an underestimation of home range size. In the example above, the red and orange points represent years one and two of tracking data from a black bear. Panel a shows the conventional estimate for the bear’s home range based on year one and does not account for autocorrelation. Panel b shows the autocorrelated kernel density estimate (AKDE) based on year one. When the second year of data is included (c-d), we can see that the conventional estimate fails to predict the year two data (c), while the AKDE predicts the appropriate proportion of future space use (d). (Figure taken from Noonan et al. 2019)



Autocorrelated Kernel Density Estimation
Dealing with telemetry error
Periodic movement models
Variograms and Model Selection


  1. M. J. Noonan, C. H. Fleming, M. A. Tucker, R. Kays, A. Harrison, M. C. Crofoot, B. Abrahms et al. “Effects of body size on estimation of mammalian area requirements.” Conservation Biology (2020).
  2. C. H. Fleming, M. J. Noonan, E. P. Medici, J. M. Calabrese, “Overcoming the challenge of small effective sample sizes in home-range estimation”, Methods in Ecology and Evolution 10:10, 1679-1689 (2019)
  3. M. J. Noonan, M. A. Tucker, C. H. Fleming, et al, “A comprehensive analysis of autocorrelation and bias in home range estimation”, Ecological Monographs, 89:2, e01344 (2019)
  4. C. H. Fleming, D. Sheldon, W. F. Fagan, P. Leimgruber, T. Mueller, D. Nandintsetseg, M. J. Noonan, K. A. Olson, E. Setyawan, A. Sianipar, J. M. Calabrese, “Correcting for missing and irregular data in home-range estimation”, Ecological Applications 28:4, 1003-1010 (2018)
  5. C. H. Fleming, J. M. Calabrese, “A new kernel-density estimator for accurate home-range and species-range area estimation”, Methods in Ecology and Evolution 8, 571-579 (2016)
  6. C. H. Fleming, W. F. Fagan, T. Mueller, K. A. Olson, P. Leimgruber, J. M. Calabrese, “Rigorous home-range estimation with movement data: A new autocorrelated kernel-density estimator”, Ecology 96:5, 1182-1188 (2015)



GPS location data are a discrete sequence of animal locations estimated at specific times and represent a sparse approximation of the animal’s unknown movement path. However, a signature of the underlying movement path may remain in the data via its autocorrelation structure. Autocorrelation in animal tracking data is an expression of the fact that locations obtained more closely in time also tend to geographically closer. The precise relationship between distance and time contains a wealth of information regarding the animal’s movement behavior and can be leveraged to make more accurate predictions.

Stay In Touch

Sign up to receive updates on our projects or send us an email

Join our mailing list

* indicates required
Contact Us